From cerha@brailcom.org Thu Sep 25 00:39:47 2003 From: Tomas Cerha To: speechd-discuss@nongnu.org Date: Sun Aug 13 12:52:45 2006 Subject: Mailing list archive Message-ID: <3F721D33.6040103@brailcom.org> Hello, I have set up aa archive and a www interface to this archive for this mailing list. You can access it at http://archives.freebsoft.org. Best Regards, Tomas Cerha. From pdm@zamazal.org Tue Oct 14 12:01:18 2003 From: Milan Zamazal To: speechd-discuss@nongnu.org Date: Sun Aug 13 12:52:45 2006 Subject: =?iso-8859-2?q?speechd_=E8te_m=EDsto_jedn=E9_te=E8ky_dv=EC?= Message-ID: <87u16c3zpt.fsf@zamazal.org> N?sleduj?c? speechd relace m? za n?sledek p?e?ten? dvou te?ek m?sto po?adovan? jedn?: % socket -c localhost 6560 set self output_module festival 216 OK OUTPUT MODULE SET set self punctuation all 205 OK PUNCTUATION SET speak 230 OK RECEIVING DATA .. . 225 OK MESSAGE QUEUED Do Festivalu skute?n? doraz? ".." m?sto spr?vn?ho ".". From philsfree@free.fr Sat Nov 8 12:41:49 2003 From: Phil's Free To: speechd-discuss@nongnu.org Date: Sun Aug 13 12:52:45 2006 Subject: Speechd/Flite/Emacs Message-ID: <1068291709.24038.93.camel@toinette.vif.argent.fr> Hi, I've just downloaded : speechd-0.1.tar.gz speechd-el-0.1.tar.gz I've succeded in compiling and launching speechd root 3247 1 0 00:58 ? 00:00:00 speechd -d -l 5 root 3248 3247 0 00:58 ? 00:00:00 /usr/local/etc/speechd/modules//flite.conf root 3250 3247 0 00:58 ? 00:00:00 /usr/local/etc/speechd/modules//epos-generic.conf root 3251 3247 0 00:58 ? 00:00:00 speechd -d -l 5 root 3252 3250 0 00:58 ? 00:00:00 /usr/local/etc/speechd/modules//epos-generic.conf root 3253 3252 0 00:58 ? 00:00:00 /usr/local/etc/speechd/modules//epos-generic.conf root 3254 3251 0 00:58 ? 00:00:00 speechd -d -l 5 root 3255 3248 0 00:58 ? 00:00:00 /usr/local/etc/speechd/modules//flite.conf root 3256 3255 0 00:58 ? 00:00:00 /usr/local/etc/speechd/modules//flite.conf But while making speechd-el, I get the following error : emacs --batch -l speechd.el -f batch-byte-compile speechd.el Loading 00debian-vars... Loading 50auctex (source)... Loading 50cxref (source)... Loading cxref... Loading 50emacspeak (source)... Loading 50gettext (source)... Loading 50nowebm (source)... Cannot open load file: queue-f Could you help me please ? Best regards and congratulations to the development team. -- Phil From buchal@brailcom.org Mon Nov 10 10:15:34 2003 From: Jan Buchal To: speechd-discuss@nongnu.org Date: Sun Aug 13 12:52:45 2006 Subject: Speechd/Flite/Emacs In-Reply-To: <1068291709.24038.93.camel@toinette.vif.argent.fr> (Phil's Free's message of "08 Nov 2003 12:41:49 +0100") References: <1068291709.24038.93.camel@toinette.vif.argent.fr> Message-ID: >>>>> "PB" == Phil's Free writes: PB> Hi, Hello, PB> emacs --batch -l speechd.el -f batch-byte-compile speechd.el Loading PB> 00debian-vars... Loading 50auctex (source)... Loading 50cxref PB> (source)... Loading cxref... Loading 50emacspeak (source)... Loading PB> 50gettext (source)... Loading 50nowebm (source)... Cannot open load PB> file: queue-f PB> Could you help me please ? Yes, I think you need install elib package. PB> Best regards and congr -- Jan Buchal Tel: (00420) 224921679 Mob: (00420) 608023021 From pdm@zamazal.org Sat Nov 8 21:56:46 2003 From: Milan Zamazal To: speechd-discuss@nongnu.org Date: Sun Aug 13 12:52:45 2006 Subject: Speechd/Flite/Emacs In-Reply-To: <1068291709.24038.93.camel@toinette.vif.argent.fr> (Phil's Free's message of "08 Nov 2003 12:41:49 +0100") References: <1068291709.24038.93.camel@toinette.vif.argent.fr> Message-ID: <873ccya8a9.fsf@zamazal.org> >>>>> "PF" == Phil's Free writes: PF> But while making speechd-el, I get the following error : PF> emacs --batch -l speechd.el -f batch-byte-compile speechd.el PF> Loading 00debian-vars... PF> Loading 50auctex (source)... PF> Loading 50cxref (source)... PF> Loading cxref... PF> Loading 50emacspeak (source)... PF> Loading 50gettext (source)... PF> Loading 50nowebm (source)... PF> Cannot open load file: queue-f PF> Could you help me please ? Install the Elib package (http://ftp.debian.org/pool/main/e/elib/). PF> Best regards and congratulations to the development team. Thanks. Regards, Milan Zamazal -- _/_\_/_ o _\_/_\_ o _/_\_/_ o _\_/_\_ o BEWARE! -<_|_|_|_><-- -<_|_|_|_><-- -<_|_|_|_><-- -<_|_|_|_><-- *Bugs* are / \ / o \ / \ o / \ / o \ / \ o approaching! From philsfree@free.fr Mon Nov 10 12:10:34 2003 From: Phil's Free To: speechd-discuss@nongnu.org Date: Sun Aug 13 12:52:45 2006 Subject: Speechd/Flite/Emacs In-Reply-To: <873ccya8a9.fsf@zamazal.org> References: <1068291709.24038.93.camel@toinette.vif.argent.fr> <873ccya8a9.fsf@zamazal.org> Message-ID: <1068462634.23446.18.camel@toinette.vif.argent.fr> Le sam 08/11/2003 ? 21:56, Milan Zamazal a ?crit : > >>>>> "PF" == Phil's Free writes: > > PF> But while making speechd-el, I get the following error : > > PF> emacs --batch -l speechd.el -f batch-byte-compile speechd.el > PF> Loading 00debian-vars... > PF> Loading 50auctex (source)... > PF> Loading 50cxref (source)... > PF> Loading cxref... > PF> Loading 50emacspeak (source)... > PF> Loading 50gettext (source)... > PF> Loading 50nowebm (source)... > PF> Cannot open load file: queue-f > > PF> Could you help me please ? > > Install the Elib package (http://ftp.debian.org/pool/main/e/elib/). > Superb ! It works very fine :-) -- Phil From rb@tertius.net.au Thu Nov 20 03:00:54 2003 From: Ricky Buchanan To: speechd-discuss@nongnu.org Date: Sun Aug 13 12:52:45 2006 Subject: Hello In-Reply-To: <1069292843.24371.ezmlm@freebsoft.org> References: <1069292843.24371.ezmlm@freebsoft.org> Message-ID: <20031120020054.GF24237@tertius.net.au> Greetings all, I'm not sure who's here but I'm subscribed now :) I'm interested in writing a Perl module to give any Perl programmer access to the Speech Dispacher API. I'm disabled myself, and it's going to take me a good long while to get anything working, but I'll do my best! Who else is here and what are you all doing? Regards, Ricky -- : Usual state: (e) None of the above. : rb@tertius.net.au http://tertius.net.au/~rb/ : Every blade of grass has an angel that bends over it and : whispers, 'Grow! Grow!' -- The Talmud From pdm@brailcom.org Thu Nov 20 09:38:31 2003 From: Milan Zamazal To: speechd-discuss@nongnu.org Date: Sun Aug 13 12:52:45 2006 Subject: Hello In-Reply-To: <20031120020054.GF24237@tertius.net.au> (Ricky Buchanan's message of "Thu, 20 Nov 2003 13:00:54 +1100") References: <1069292843.24371.ezmlm@freebsoft.org> <20031120020054.GF24237@tertius.net.au> Message-ID: <87r803zb7c.fsf@zamazal.org> >>>>> "RB" == Ricky Buchanan writes: RB> Who else is here and what are you all doing? I'm here :-), one of the Speech Dispatcher team people. I focus on speechd-el (Emacs speech support using Speech Dispatcher) and Festival. Milan Zamazal -- I think any law that restricts independent use of brainpower is suspect. -- Kent Pitman in comp.lang.lisp From cerha@brailcom.org Thu Nov 20 18:15:21 2003 From: Tomas Cerha To: speechd-discuss@nongnu.org Date: Sun Aug 13 12:52:45 2006 Subject: Speech Dispacher + Perl In-Reply-To: <87znerzbie.fsf@zamazal.org> References: <20031119134449.GA24141@tertius.net.au> <20031119153735.GB867@localhost.localdomain> <20031120023427.GB7008@tertius.net.au> <87znerzbie.fsf@zamazal.org> Message-ID: <3FBCF6A9.2020504@brailcom.org> Milan Zamazal wrote: >>>>>>"RB" == Ricky Buchanan writes: > > > RB> The main problem I'll probably have is getting the module to > RB> talk to the C library stuff in SpeechD. > > You might want to talk to Speech Dispatcher directly, via a socket > communication, using the SSIP protocol described in the Speech > Dispatcher documentation. (I personally prefer language native > libraries over wrapping C libraries, for headache avoidance reasons.:-) You can also have a look at Python interface, which is included in current speechd sources. It is very basic, but it works, so you can use it as an inspiration for writing the Perl interface. Best regards, Tomas Cerha. From cerha@brailcom.org Thu Nov 20 18:33:35 2003 From: Tomas Cerha To: speechd-discuss@nongnu.org Date: Sun Aug 13 12:52:45 2006 Subject: Hello In-Reply-To: <20031120020054.GF24237@tertius.net.au> References: <1069292843.24371.ezmlm@freebsoft.org> <20031120020054.GF24237@tertius.net.au> Message-ID: <3FBCFAEF.1090704@brailcom.org> Ricky Buchanan wrote: > I'm interested in writing a Perl module to give any Perl > programmer access to the Speech Dispacher API. Hello Ricky, If you want me to create a CVS repository for the Perl interface, we can run it as a separate Free(b)soft project. We have quite a nice framework for our projects, so that you could make a use of on-line documentation generation system, website, mailing-lists etc. Best regards, Tomas Cerha . From philsfree@free.fr Fri Nov 21 01:20:37 2003 From: Phil's Free To: speechd-discuss@nongnu.org Date: Sun Aug 13 12:52:45 2006 Subject: Hello In-Reply-To: <3FBCFAEF.1090704@brailcom.org> References: <1069292843.24371.ezmlm@freebsoft.org> <20031120020054.GF24237@tertius.net.au> <3FBCFAEF.1090704@brailcom.org> Message-ID: <1069374037.24626.4.camel@toinette.vif.argent.fr> Le jeu 20/11/2003 ? 18:33, Tomas Cerha a ?crit : > Ricky Buchanan wrote: > > I'm interested in writing a Perl module to give any Perl > > programmer access to the Speech Dispacher API. > > Hello Ricky, > > If you want me to create a CVS repository for the Perl interface, we can run it > as a separate Free(b)soft project. We have quite a nice framework for our > projects, so that you could make a use of on-line documentation generation > system, website, mailing-lists etc. > Good idea ! I'm also interested in a Perl interface for Speech Dispatcher. I belong to the BigLux project and have written a very basic phonetizer in Perl for the french language. Regards. -- Phil Buch From rb@tertius.net.au Mon Nov 24 22:31:18 2003 From: Ricky Buchanan To: speechd-discuss@nongnu.org Date: Sun Aug 13 12:52:45 2006 Subject: Hello In-Reply-To: <1069374037.24626.4.camel@toinette.vif.argent.fr> References: <1069292843.24371.ezmlm@freebsoft.org> <20031120020054.GF24237@tertius.net.au> <3FBCFAEF.1090704@brailcom.org> <1069374037.24626.4.camel@toinette.vif.argent.fr> Message-ID: <20031124213118.GA18517@tertius.net.au> > Le jeu 20/11/2003 ? 18:33, Tomas Cerha a ?crit : > > Ricky Buchanan wrote: > > > I'm interested in writing a Perl module to give any Perl > > > programmer access to the Speech Dispacher API. > > If you want me to create a CVS repository for the Perl interface, we can run it > > as a separate Free(b)soft project. We have quite a nice framework for our > > projects, so that you could make a use of on-line documentation generation > > system, website, mailing-lists etc. Not really interested in all that stuff. I have my own framework worked out over many years, and I'm used to it. And due to the disability/cognitive problems it's a hell of a challenge to use something different and would probably not allow me to actually do any Real Work because I'd get caught up grappling with somebody else's interface. I hope that doesn't cause problems ... Easiest for me would just be to email privately with Phil and anybody else interested in helping - I'll set up speechd-perl@tertius.net.au if more than 2-3 others become interested but at the moment a mailing list is overkill I think. Also I routinely stick all my code up at http://tertius.net.au/~rb/code/, part of my makefiles for my perl projects automatically does this when I roll a new distribution, so there's no work involved as there would be trying to get stuff uploaded to somebody else's website. Online documentation system - I have no idea what this means. Perl modules are internally documented using a POD syntem, you type perldoc and it extracts the documentation from the module and shows it to you in a slightly formatted way, usually the install process also converts this to a manpage automatically. And since it's actually *in* the sourcefile, it serves as internal documentation too as the Phil's Free a ?crit : > Good idea ! > I'm also interested in a Perl interface for Speech Dispatcher. > I belong to the BigLux project and have written a very basic phonetizer > in Perl for the french language. I'm not familiar with BigLux - what is that project? What parts of this project are you interested in, and do you have any use for a perl interface to speechd in mind already, or are you working in the abstract? Regards, Ricky -- : Usual state: (e) None of the above. : rb@tertius.net.au http://tertius.net.au/~rb/ : I'm right here. This is who I really am. I'm not pretending. : -- Orson Scott Card, 'Ender's Shadow' From rb@tertius.net.au Mon Nov 24 22:36:29 2003 From: Ricky Buchanan To: speechd-discuss@nongnu.org Date: Sun Aug 13 12:52:45 2006 Subject: Speech Dispacher + Perl In-Reply-To: <3FBCF6A9.2020504@brailcom.org> References: <20031119134449.GA24141@tertius.net.au> <20031119153735.GB867@localhost.localdomain> <20031120023427.GB7008@tertius.net.au> <87znerzbie.fsf@zamazal.org> <3FBCF6A9.2020504@brailcom.org> Message-ID: <20031124213629.GB18517@tertius.net.au> Tomas Cerha wrote: > Milan Zamazal wrote: > > You might want to talk to Speech Dispatcher directly, via a socket > > communication, using the SSIP protocol described in the Speech > > Dispatcher documentation. (I personally prefer language native > > libraries over wrapping C libraries, for headache avoidance reasons.:-) > You can also have a look at Python interface, which is included in current > speechd sources. It is very basic, but it works, so you can use it as an > inspiration for writing the Perl interface. Luckily, the two interfaces already existant are in the two languages I can read best, besides Perl, so I plan to read them both. Have read the C interface and the info document describing SSIP already and it looks quite easy to use. I am sure I'll be proved wrong but it looks to me like it'll actually be a very simple project to knock up an interface for the thing! Yay! BTW, you said the Py interface was in the current sources - where exactly? It's not in the speech-dispatcher_0.1.orig.tar.gz tarball I downloaded. Did you mean hidden in the CVS repository? If so, could you give me exact instructions for grabbing it, something I can paste into a get_python_interface.sh script? I really am TOTALLY CRAPPY at CVS. Unfortunately. I keep trying to learn the damm thing and it keeps defeating me. The fact that one of my best friends wrote the O'Reilly book on the topic doesn't help, I'm sure!! Regards, Ricky -- : Usual state: (e) None of the above. : rb@tertius.net.au http://tertius.net.au/~rb/ : I'm right here. This is who I really am. I'm not pretending. : -- Orson Scott Card, 'Ender's Shadow' From cerha@brailcom.org Mon Nov 24 23:10:38 2003 From: Tomas Cerha To: speechd-discuss@nongnu.org Date: Sun Aug 13 12:52:45 2006 Subject: Hello In-Reply-To: <20031124213118.GA18517@tertius.net.au> References: <1069292843.24371.ezmlm@freebsoft.org> <20031120020054.GF24237@tertius.net.au> <3FBCFAEF.1090704@brailcom.org> <1069374037.24626.4.camel@toinette.vif.argent.fr> <20031124213118.GA18517@tertius.net.au> Message-ID: <3FC281DE.2000009@brailcom.org> Ricky Buchanan wrote: >>>If you want me to create a CVS repository for the Perl interface, we can run it >>>as a separate Free(b)soft project. We have quite a nice framework for our >>>projects, so that you could make a use of on-line documentation generation >>>system, website, mailing-lists etc. > > Not really interested in all that stuff. I have my own framework worked > out over many years, and I'm used to it. And due to the > disability/cognitive problems it's a hell of a challenge to use > something different and would probably not allow me to actually do any > Real Work because I'd get caught up grappling with somebody else's > interface. I hope that doesn't cause problems ... Our framework is actually designed for impaired people, however I perfectly understand your reasons. > Easiest for me would just be to email privately with Phil and anybody > else interested in helping - I'll set up speechd-perl@tertius.net.au if > more than 2-3 others become interested but at the moment a mailing list > is overkill I think. I'd be definitely interested. I was going to write a Perl interface myself, however I haven't done so yet because of lack of time and demand from outside... I think I might help. > Also I routinely stick all my code up at > http://tertius.net.au/~rb/code/, part of my makefiles for my perl > projects automatically does this when I roll a new distribution, so > there's no work involved as there would be trying to get stuff uploaded > to somebody else's website. No worries, we will just link it from our website... > Online documentation system - I have no idea what this means. Perl > modules are internally documented using a POD syntem, you type > perldoc That's it. We just keep an on-line html versions of documentation for all our projects. It is simply regenerated each time you commit the sources to CVS. That's all :-) Best regards, Tomas. From rb@tertius.net.au Mon Nov 24 23:22:27 2003 From: Ricky Buchanan To: speechd-discuss@nongnu.org Date: Sun Aug 13 12:52:45 2006 Subject: Hello In-Reply-To: <3FC281DE.2000009@brailcom.org> References: <1069292843.24371.ezmlm@freebsoft.org> <20031120020054.GF24237@tertius.net.au> <3FBCFAEF.1090704@brailcom.org> <1069374037.24626.4.camel@toinette.vif.argent.fr> <20031124213118.GA18517@tertius.net.au> <3FC281DE.2000009@brailcom.org> Message-ID: <20031124222227.GA20314@tertius.net.au> Tomas Cerha wrote: > Ricky Buchanan wrote: > >>>If you want me to create a CVS repository for the Perl interface, we can run it > >>>as a separate Free(b)soft project. We have quite a nice framework for our > >>>projects, so that you could make a use of on-line documentation generation > >>>system, website, mailing-lists etc. > > Not really interested in all that stuff. I have my own framework worked > > out over many years, and I'm used to it. And due to the > > disability/cognitive problems it's a hell of a challenge to use > > something different and would probably not allow me to actually do any > > Real Work because I'd get caught up grappling with somebody else's > > interface. I hope that doesn't cause problems ... > Our framework is actually designed for impaired people, however I perfectly > understand your reasons. Isn't it designed for those with *visual* impairments though, not physical + cognitive like mine? I'd be interested (perhaps offlist?) in how you've designed the interface to be accessible and who for ... Regards, Ricky -- : Usual state: (e) None of the above. : rb@tertius.net.au http://tertius.net.au/~rb/ : Sometimes there's no point in giving up. : -- Louis Wu, 'The Ringworld Engineers' From cerha@brailcom.org Tue Nov 25 00:10:28 2003 From: Tomas Cerha To: speechd-discuss@nongnu.org Date: Sun Aug 13 12:52:45 2006 Subject: Speech Dispacher + Perl In-Reply-To: <20031124213629.GB18517@tertius.net.au> References: <20031119134449.GA24141@tertius.net.au> <20031119153735.GB867@localhost.localdomain> <20031120023427.GB7008@tertius.net.au> <87znerzbie.fsf@zamazal.org> <3FBCF6A9.2020504@brailcom.org> <20031124213629.GB18517@tertius.net.au> Message-ID: <3FC28FE4.1010509@brailcom.org> Ricky Buchanan wrote: > BTW, you said the Py interface was in the current sources - where > exactly? It's not in the speech-dispatcher_0.1.orig.tar.gz tarball I > downloaded. Did you mean hidden in the CVS repository? Yes, in the CVS it's the /src/python directory. > If so, could > you give me exact instructions for grabbing it, something I can paste > into a get_python_interface.sh script? wget http://cvs.freebsoft.org/repository/speechd/src/python.tar.gz?view=tar Regards, Tomas. From pdm@brailcom.org Tue Nov 25 09:11:49 2003 From: Milan Zamazal To: speechd-discuss@nongnu.org Date: Sun Aug 13 12:52:45 2006 Subject: Speech Dispacher + Perl In-Reply-To: <20031124213629.GB18517@tertius.net.au> (Ricky Buchanan's message of "Tue, 25 Nov 2003 08:36:29 +1100") References: <20031119134449.GA24141@tertius.net.au> <20031119153735.GB867@localhost.localdomain> <20031120023427.GB7008@tertius.net.au> <87znerzbie.fsf@zamazal.org> <3FBCF6A9.2020504@brailcom.org> <20031124213629.GB18517@tertius.net.au> Message-ID: <87u14samuy.fsf@zamazal.org> >>>>> "RB" == Ricky Buchanan writes: RB> the two interfaces already existant are in the two languages Just a small correction -- there are three interfaces for three languages: C, Elisp, and Python. BTW, if anybody is interested in a Common Lisp interface, let me know. Milan Zamazal -- _/_\_/_ o _\_/_\_ o _/_\_/_ o _\_/_\_ o BEWARE! -<_|_|_|_><-- -<_|_|_|_><-- -<_|_|_|_><-- -<_|_|_|_><-- *Bugs* are / \ / o \ / \ o / \ / o \ / \ o approaching! From hanke@volny.cz Sun Dec 7 23:39:59 2003 From: Hynek Hanke To: speechd-discuss@nongnu.org Date: Sun Aug 13 12:52:45 2006 Subject: Document about SSIP Message-ID: <20031207223959.GA3161@localhost.localdomain> Hi all, I've finally managed to assemble a document about the Speech Synthesis Internet Protocol. It's independent of Speech Dispatcher (it doesn't even refer to it -- which I don't know if it's a good thing). It's mostly splitted from speechd.texi, with some corrections. But some parts are also newly written, especially in the section Introduction, because I found that the corresponding parts in speechd.texi are too closely tight to Speech Dispatcher that I cannot use them in a document that speaks purely of SSIP. Please comment on it, or even better, make changes directly in the document. I'm not sure if the extent of the introduction section is ok or it's too short for people to understand. I'm also not sure about the contact adress on the front page. It's kontakt@brailcom.cz (because Brailcom, o.p.s. is stated as the author) but it should probably be something other. It doesn't build automatically, because I'm not sure about its status in the Info directory. But you can easily generate a PDF through texi2pdf ssip.texi in speechd/doc in the source tree or info by: makeinfo ssip.texi Best Regards, Hynek Hanke From pdm@freebsoft.org Wed Dec 10 18:56:29 2003 From: Milan Zamazal To: speechd-discuss@nongnu.org Date: Sun Aug 13 12:52:45 2006 Subject: speechd-el 0.2 release candidate available Message-ID: <87r7zcsgj6.fsf@zamazal.org> speechd-el 0.2 release candidate is available at http://www.freebsoft.org/pub/projects/speechd-el/speechd-el-0.2rc1.tar.gz The corresponding experimental Debian packages are available in http://www.freebsoft.org/pub/projects/speechd-el/deb/ Please test the release candidate if you can and let me know if anything is wrong. * What is new in the 0.2 version? - Bug fixes. - read-event prompts are spoken now. - Spelling support added. Thanks, Milan Zamazal -- http://www.zamazal.org From pdm@freebsoft.org Mon Dec 15 08:34:10 2003 From: Milan Zamazal To: speechd-discuss@nongnu.org Date: Sun Aug 13 12:52:45 2006 Subject: speechd-el 0.2 released Message-ID: <87llpe1qml.fsf@zamazal.org> speechd-el 0.2 released ======================= The Free(b)soft project is pleased to announce speechd-el 0.2. * What is speechd-el? speechd-el is an Emacs client to Speech Dispatcher, providing a complex speech interface to Emacs, focused especially on (but not limited to) the blind and visually impaired users. It allows the user to work with Emacs without looking on the screen, using the speech output produced by the synthesizers supported in Speech Dispatcher. Key speechd-el features are: - Automated speaking of most Emacs messages and actions, without necessity to customize each particular Emacs function to make it speak. - Most built-in and external Emacs packages can produce speech output immediately, without any special support. - Almost no change of the standard Emacs behavior, no interference with user settings. - Highly configurable. - Full cooperation with Speech Dispatcher. - Small code size. - Free Software, distributed under the GPL. Part of the speechd-el package is an Elisp library for communication with Speech Dispatcher, allowing you to write your own Emacs speech output extensions and your own Emacs interfaces to Speech Dispatcher. * What is new in the 0.2 version? - Bug fixes. - read-event prompts are spoken now. - Spelling support added. - Choosing widget values improved. See the NEWS file for more details. * Where to get it? You can get the distribution tarball of the released version from http://www.freebsoft.org/pub/projects/speechd-el/speechd-el-0.2.tar.gz . The home page of the project is http://www.freebsoft.org/speechd-el . From rb@tertius.net.au Tue Dec 23 16:22:39 2003 From: Ricky Buchanan To: speechd-discuss@nongnu.org Date: Sun Aug 13 12:52:45 2006 Subject: First perl release! Message-ID: <20031223152239.GB22943@tertius.net.au> Given the previous message, it is untested, even the code, but apart from that it works *wry grin* Check out http://tertius.net.au/~rb/code/ Go into Speech-SpeechDispacher or whatever its called, then check out the stuff. Hope you like it. I wish I had some decent speaker-independant voice rec to go with this!!! Pipe dreaming, Ricky -- : Usual state: (e) None of the above. : rb@tertius.net.au http://tertius.net.au/~rb/ : That which is static and repetitive is boring. That which is : dynamic and random is confusing. In between lies art. : -- John A Locke From rb@tertius.net.au Tue Dec 23 16:21:22 2003 From: Ricky Buchanan To: speechd-discuss@nongnu.org Date: Sun Aug 13 12:52:45 2006 Subject: No output! Message-ID: <20031223152122.GA22943@tertius.net.au> 1. Tried testing speech dispacher today for the first time, strangely. Logfile looks OK: [Wed Dec 24 02:07:26 2003] speechd: Activity on fd 8 ... [Wed Dec 24 02:07:26 2003] speechd: Read 4 bytes from client on fd 8 [Wed Dec 24 02:07:26 2003] speechd: Activity on fd 8 ... [Wed Dec 24 02:07:26 2003] speechd: Read 3 bytes from client on fd 8 [Wed Dec 24 02:07:26 2003] speechd: Finishing data [Wed Dec 24 02:07:26 2003] speechd: Switching back to command mode... [Wed Dec 24 02:07:26 2003] speechd: New buf is now: |Hello, I'm a Speech Dispatcher communication script test! Ricky Buchanan wrote me. B.| [Wed Dec 24 02:07:26 2003] speechd: Posting on semaphore. [Wed Dec 24 02:07:26 2003] speechd: Message in queue p2 [Wed Dec 24 02:07:26 2003] speechd: Module set. [Wed Dec 24 02:07:26 2003] speechd: Module speak! [Wed Dec 24 02:07:26 2003] speechd: Message sent to output module [Wed Dec 24 02:07:29 2003] speechd: Activity on fd 8 ... [Wed Dec 24 02:07:29 2003] speechd: Read 6 bytes from client on fd 8 [Wed Dec 24 02:07:29 2003] speechd: Bye received. [Wed Dec 24 02:07:29 2003] speechd: Removing client on fd 8 [Wed Dec 24 02:07:29 2003] speechd: Tagging client as inactive in settings [Wed Dec 24 02:07:29 2003] speechd: Removing client from the fd->uid table. [Wed Dec 24 02:07:29 2003] speechd: Closing clients file descriptor 8 [Wed Dec 24 02:07:29 2003] speechd: Connection closed Festival is running. Festival works when tested directly. Anything ELSE to test??? *looks lost* Regards, Ricky -- : Usual state: (e) None of the above. : rb@tertius.net.au http://tertius.net.au/~rb/ : That which is static and repetitive is boring. That which is : dynamic and random is confusing. In between lies art. : -- John A Locke From rb@tertius.net.au Tue Dec 23 20:38:35 2003 From: Ricky Buchanan To: speechd-discuss@nongnu.org Date: Sun Aug 13 12:52:45 2006 Subject: No output! In-Reply-To: <20031223202739.GD1490@volny.cz> References: <20031223152122.GA22943@tertius.net.au> <20031223202739.GD1490@volny.cz> Message-ID: <20031223193835.GA16713@tertius.net.au> Hynek Hanke wrote: > On Wed, Dec 24, 2003 at 02:21:22AM +1100, Ricky Buchanan wrote: > > Logfile looks OK: > > [...] > > Festival is running. > > Festival works when tested directly. > > Anything ELSE to test??? > Please try to test if you are allowed to connect to festival > server. Try > telnet localhost 1314 Works perfectly: [rb@mycroft] ~/logs> telnet localhost 1314 Trying 127.0.0.1... Connected to localhost. Escape character is '^X^]'. (SayText "testing 1 2 3 ") LP # ft_StUfF_keyOK (quit) ER Connection closed by foreign host. [rb@mycroft] ~/logs> And yes, the utterance is heard too :) > If telnet seems to work, you can get additional information from > the festival output module by enabling debug logging. You can do > so in etc/speechd/modules/festival.conf (or some similar path to > festival.conf). Please repeat the problem and send us this (complete) > log so that we can look at what went wrong. Done. log is attached. Weird thing is that the soundfile is correctly CREATED and saved and can be easily played with $ esdplay debug-festival-0.snd And yes, festival is set up to be worked properly with Festival! From buchal@brailcom.org Tue Dec 23 21:00:39 2003 From: Jan Buchal To: speechd-discuss@nongnu.org Date: Sun Aug 13 12:52:45 2006 Subject: No output! In-Reply-To: <20031223193835.GA16713@tertius.net.au> (Ricky Buchanan's message of "Wed, 24 Dec 2003 06:38:35 +1100") References: <20031223152122.GA22943@tertius.net.au> <20031223202739.GD1490@volny.cz> <20031223193835.GA16713@tertius.net.au> Message-ID: <87oetzmhi0.fsf@brailcom.org> >>>>> "RB" == Ricky Buchanan writes: RB> [rb@mycroft] ~/logs> telnet localhost 1314 Trying 127.0.0.1... RB> Connected to localhost. Escape character is '^X^]'. (SayText "testing RB> 1 2 3 ") LP # ft_StUfF_keyOK (quit) ER Connection RB> closed by foreign host. [rb@mycroft] ~/logs> I do not know why, but is necessary put as FIRST line in to /etc/hosts 127.0.0.1 localhost Please try this and let know. -- Jan Buchal From hanke@volny.cz Wed Dec 24 00:15:14 2003 From: Hynek Hanke To: speechd-discuss@nongnu.org Date: Sun Aug 13 12:52:45 2006 Subject: No output! In-Reply-To: <87oetzmhi0.fsf@brailcom.org> References: <20031223152122.GA22943@tertius.net.au> <20031223202739.GD1490@volny.cz> <20031223193835.GA16713@tertius.net.au> <87oetzmhi0.fsf@brailcom.org> Message-ID: <20031223231514.GC1571@volny.cz> On Tue, Dec 23, 2003 at 09:00:39PM +0100, Jan Buchal wrote: > >>>>> "RB" == Ricky Buchanan writes: > RB> [rb@mycroft] ~/logs> telnet localhost 1314 Trying 127.0.0.1... > RB> Connected to localhost. Escape character is '^X^]'. (SayText "testing > RB> 1 2 3 ") LP # ft_StUfF_keyOK (quit) ER Connection > RB> closed by foreign host. [rb@mycroft] ~/logs> > I do not know why, but is necessary put as FIRST line in to /etc/hosts > 127.0.0.1 localhost > Please try this and let know. I don't think this is necessary *after* checking that telnet to the port works correctly. If telnet works, Speech Dispatcher also has to work because the connection originates on the same host. Have a nice Christmars all, Hynek Hanke From hanke@volny.cz Fri Dec 26 15:21:23 2003 From: Hynek Hanke To: speechd-discuss@nongnu.org Date: Sun Aug 13 12:52:45 2006 Subject: Speakup & Speech Dispatcher In-Reply-To: <3FD97DB6.5010601@brailcom.org> References: <3FD97DB6.5010601@brailcom.org> Message-ID: <20031226142123.GA1661@volny.cz> > Now the real reason I am writing. I have just finished my software > synthesizer interface driver for speakup and have been trying to > decide how to do the interfacing to software synths. I either have to > write a middleware program to drive various software synths myself or > find some other method and speech dispatcher came to mind. Hi Kirk, I'm very happy to hear from you again. Your driver for communication with user-space applications sounds very well. Be sure we would really like to have some interfacing from Speakup to Speech Dispatcher! I think we have to write some simple daemon application that would read the commands from /dev/softsynth and then send them to Speech Dispatcher in SSIP. It would act as a Speech Dispatcher client. This shouldn't be very difficult to write and I think I could do it myself relatively soon. > On a read() it will pass > all output from the kernel interspersed with pseudo synth control > strings based on the Doubletalk command set kind of. I say kind of > because it isn't exactly the same I use values 0-9 for pitch instead > of 0-99, other than that it's identical. I think maybe it would be easier and clearer to pass the commands directly in SSIP. It would also make the task of writing the interface daemon easier. Instead of creating your own command set. Also, I think that the Doubletalk protocol lacks some of the features supported by SSIP, mainly the priority system. Software synthesis is capable of doing more than the simple hardware devices and it would be great if we could use the most of what current technology offers (or just leave the doors open). Even if there is no direct use for priorities & history & sound icons right now in Speakup, it might come later. You told me that you are thinking about some application-specific customization of Speakup. When you implement it, Speakup will be able to distinguish different messages as less or more important and with the Doubletalk commands, you'll have to invent yet another command set. I'd like to know what do you think about this? Well, obviously, I don't know if there isn't any major obstacle to using SSIP or it's derivative. Do you think it would be hard to replace the current Doubetalk commands with SSIP commands or it's just a matter of rewriting the strings here and there? I'm CC-ing this message to speechd@freebsoft.org. Maybe it would be easier to communicate here. You are invited to join by sending a mail to speechd-subscribe@freebsoft.org . Thanks for your message @ nice rest of the Christmas holiday, Hynek From pdm@freebsoft.org Mon Dec 29 19:46:06 2003 From: Milan Zamazal To: speechd-discuss@nongnu.org Date: Sun Aug 13 12:52:45 2006 Subject: No output! In-Reply-To: <20031223193835.GA16713@tertius.net.au> (Ricky Buchanan's message of "Wed, 24 Dec 2003 06:38:35 +1100") References: <20031223152122.GA22943@tertius.net.au> <20031223202739.GD1490@volny.cz> <20031223193835.GA16713@tertius.net.au> Message-ID: <87ad5bxy1c.fsf@zamazal.org> Clearly, there is a problem opening the audio device. Does the speechd process have the necessary permissions to access the device? Isn't it blocked by esd? Speech Dispatcher uses flite audio library for its sound output, if you have installed flite, does it work with your audio setup? Regards, Milan Zamazal -- Why waste keystrokes when you can waste CPU cycles instead? Karl Fogel From hanke@volny.cz Mon Dec 29 21:22:39 2003 From: Hynek Hanke To: speechd-discuss@nongnu.org Date: Sun Aug 13 12:52:45 2006 Subject: No output! In-Reply-To: <87ad5bxy1c.fsf@zamazal.org> References: <20031223152122.GA22943@tertius.net.au> <20031223202739.GD1490@volny.cz> <20031223193835.GA16713@tertius.net.au> <87ad5bxy1c.fsf@zamazal.org> Message-ID: <20031229202239.GB898@volny.cz> On Mon, Dec 29, 2003 at 07:46:06PM +0100, Milan Zamazal wrote: > Clearly, there is a problem opening the audio device. Does the speechd > process have the necessary permissions to access the device? Isn't it > blocked by esd? Speech Dispatcher uses flite audio library for its > sound output, if you have installed flite, does it work with your audio > setup? We already know what's the problem. It is esd running and blocking /dev/dsp. I've written a longer reply, but I misposted it only to Ricky. I'll definitaly shift it up in my priority list, because the Flite audio library is really annoying. Regards, Hynek Hanke From rb@tertius.net.au Mon Jan 5 07:05:27 2004 From: Ricky Buchanan To: speechd-discuss@nongnu.org Date: Sun Aug 13 12:52:45 2006 Subject: Audio server(s) support in Speech Dispatcher (Re: No output!) In-Reply-To: <20040104202732.GA10875@volny.cz> Message-ID: <20040104202805.GA11750@tertius.net.au> Hynek Hanke wrote: > 3) NAS (Network Audio System) > It does a similar thing to ESD, but NAS is much more powerful. > It supports synchronization and real-time stopping. It has a nice > C library for communication with it and easy support for volume > setting. Possible reason for the non-use is a MisNaming Problem. I never even *looked* at NAS because I assumed it was something to do with networking sound between different boxen. The idea it'd do what ESD does never even occurred to me! > Ricky, don't you want to consider NAS for your work? I think Yes, if it does what ESD does then I *definitely* want to look at it and will do so ASAP. > As for replacing the Flite sound library in Speech Dispatcher > with something better, I thing the NAS library looks good, > but it would require users to install yet an additional application. > If this is not what we want, I'd recommend to remain waiting > for something suitable (liba(i)o is not currently). I don't think installing applications is a problem. For most people, who use RedHat, Debian, etc., then installing another application just means typing and as long as the documentation for how to make the two cooperate is good documentation, then I'd go with it ASAP. Regards, Ricky -- : Usual state: (e) None of the above. : rb@tertius.net.au http://tertius.net.au/~rb/ : I find that the harder I work, the more luck I seem to have. : -- Thomas Jefferson From rb@tertius.net.au Mon Jan 5 07:06:30 2004 From: Ricky Buchanan To: speechd-discuss@nongnu.org Date: Sun Aug 13 12:52:45 2006 Subject: Audio server(s) support in Speech Dispatcher (Re: No output!) Message-ID: <20040105060630.GB30455@tertius.net.au> [NOTE: From Hynek! Forwarded by Ricky because he sent it to the wrong address. Can we all point and laugh now? *wink* ] ----- Begin forwarded message ----- On Wed, Dec 24, 2003 at 09:22:06AM +1100, Ricky Buchanan wrote: > Hynek Hanke wrote: > > now I understand the problem. It seems everything works correctly > > except the resulting wave isn't played because /dev/dsp isn't > > accessible as it's blocked by esd. I spent some time looking for some decent sound system and and sound library in GNU/Linux. The situation as it is today is quite sad. On the other side, Speech Dispatcher can be made to work in an environment where multiple sound sources access the soundcard simultaneously without any modifications in the code. Here is a more detailed description: 1) OSS (Open Audio System) Curently Speech Dispatcher uses oss output through some Flite sound library that writes directly to /dev/dsp. Any other application with sound output are blocked, because OSS allows only one client to access soundcard at once. So it's better to make OSS accessible only to one of the sound daemons listed bellow. 2) Alsa I don't know anything about it. But there are people on this mailing list who do. Please could someone post more information about it? Does it support synchronization and immediate stopping? Does it have some decent interface library? 2) ESD (Enlightened Sound Daemon) ESD is a daemon application trying to solve this issue by doing similar things that Speech Dispatcher does to software synthesis. It reads the streams from various clients, mixes them and sends them to /dev/dsp (which is being blocked by esd). The old-style applications that access /dev/dsp directly doesn't work when ESD is running. However, there exists a program called esddsp, that links the old-style programs with libesddsp at runtime and rewrites all calls to /dev/dsp with calls to ESD. This way, the application things it accesses /dev/dsp, but in reallity, it's talking to ESD and everything works just fine. Speech Dispatcher can work with this (tested), but with severe limitations... Unfortunatelly, if I understand things correctly (which is hard to say, because there is virtually *none* documentation to ESD), ESD in it's current state can't fully work with Speech Dispatcher because it doesn't provide any means of synchronization and the stop command is *very* inacurate (like several seconds, while in Speech Dispatcher we care about *fractions* of a second in many situations). In general, ESD doesn't look as a very good choice to me. I don't really understand why it's so widely used. 3) NAS (Network Audio System) It does a similar thing to ESD, but NAS is much more powerful. It supports synchronization and real-time stopping. It has a nice C library for communication with it and easy support for volume setting. Old-style applications that open directly /dev/dsp can be made to work with it through audiooss (a separate package) that does exactly the same as esddsp does to esd -- it maps calls to /dev/dsp to communication with NAS. Speech Dispatcher also works with NAS through this method and I didn't notice any problem during testing it. You have to install NAS and audiooss, start NAS, then execute audiooss speechd -s or audiooss speechd or you can modify your init.d/speech-dispatcher script to do it for you. In XMMS, ESD support is included by default, while NAS must be installed additionally. I don't understand this either, but NAS is much less used. -- 4) libao interface library libao is an interface library designed to talk to various sound systems, but it's capabilities are highly limited. No synchronization, no stopping. Without writing some middleware code, this is useless for real time sound output. 5) libaio interface library This is a new project, in it's early stage of developement, that aims to provide interface to the different sound systems and provide the functions necessary to real time sound output, like synchronization. Unfortunately, it currently only has a driver for Alsa, so it doesn't solve our problem very much. But the developement seems to be active, so maybe something will come out of it. Resume ====== Ricky, don't you want to consider NAS for your work? I think it's a better alternative (there is also a lot of complaints about ESD on internet). If not, you might want to try running speechd as esddsp speechd -s and see if it's limited functioning is enough for you. I don't think we could archieve anything better with ESD :( As for replacing the Flite sound library in Speech Dispatcher with something better, I thing the NAS library looks good, but it would require users to install yet an additional application. If this is not what we want, I'd recommend to remain waiting for something suitable (liba(i)o is not currently). Looking for your comments! Regards, Hynek Hanke ----- End forwarded message ----- From pdm@freebsoft.org Mon Jan 5 10:55:42 2004 From: Milan Zamazal To: speechd-discuss@nongnu.org Date: Sun Aug 13 12:52:45 2006 Subject: Audio server(s) support in Speech Dispatcher (Re: No output!) In-Reply-To: <20040105060630.GB30455@tertius.net.au> (Ricky Buchanan's message of "Mon, 5 Jan 2004 17:06:30 +1100") References: <20040105060630.GB30455@tertius.net.au> Message-ID: <87n092oh29.fsf@zamazal.org> > 2) Alsa > I don't know anything about it. But there are people on this > mailing list who do. Please could someone post more information > about it? Does it support synchronization and immediate > stopping? Does it have some decent interface library? AFAIK, ALSA doesn't have any real limits (remember, it can be used as a fully functional backend by audio libraries), actually it's quite powerful. But ALSA is basically a Linux audio driver, no way a general purpose audio library, so its direct use should be avoided in portable code. It can serve perfectly as one of Linux backends in an audio library, but Speech Dispatcher should use higher level means for its audio output. If NAS serves the needs, use it. Regards, Milan Zamazal -- http://www.zamazal.org From cerha@brailcom.org Mon Jan 5 11:43:29 2004 From: Tomas Cerha To: speechd-discuss@nongnu.org Date: Sun Aug 13 12:52:45 2006 Subject: Audio server(s) support in Speech Dispatcher (Re: No output!) In-Reply-To: <20040104202805.GA11750@tertius.net.au> References: <20040104202805.GA11750@tertius.net.au> Message-ID: <3FF93FD1.3060809@brailcom.org> >>As for replacing the Flite sound library in Speech Dispatcher >>with something better, I thing the NAS library looks good, >>but it would require users to install yet an additional application. >>If this is not what we want, I'd recommend to remain waiting >>for something suitable (liba(i)o is not currently). > > > I don't think installing applications is a problem. For most people, > who use RedHat, Debian, etc., then installing another application just > means typing > > Yes, but for people who are currently using ESD it might be a problem since ESD and NAS can't be used simultaneously. And migration to NAS requires some effort... And there might be applications that don't have NAS support at all. But as we need something now and ESD seems not to be suitable, we probably have to force people to replace ESD with something suitable. However I would not require NAS, since nowadays there are many sondcards with hardware support for multiple inputs on one DSP device, so we should also support direct DSP output. I think there are two possibilities: * using an interface library that is able at least of NAS and DSP output (maybe the NAS library?). * using direct DSP output and relying on audiooss. Than the message to the user would be: Either use a souncard with multiple channels or use NAS. ESD is unfortunately not suitable. But I also don't have a clear idea how might this be influenced by ALSA... Best Regards, Tomas. From hanke@volny.cz Mon Jan 5 14:54:40 2004 From: Hynek Hanke To: speechd-discuss@nongnu.org Date: Sun Aug 13 12:52:45 2006 Subject: Audio server(s) support in Speech Dispatcher (Re: No output!) In-Reply-To: <3FF93FD1.3060809@brailcom.org> References: <20040104202805.GA11750@tertius.net.au> <3FF93FD1.3060809@brailcom.org> Message-ID: <20040105135440.GA1087@volny.cz> On Mon, Jan 05, 2004 at 11:43:29AM +0100, Tomas Cerha wrote: > >>As for replacing the Flite sound library in Speech Dispatcher > >>with something better, I thing the NAS library looks good, > >>but it would require users to install yet an additional application. > >>If this is not what we want, I'd recommend to remain waiting > >>for something suitable (liba(i)o is not currently). > > Yes, but for people who are currently using ESD it might be a problem since ESD > and NAS can't be used simultaneously. Actually, I heard you can let ESD talk to NAS and thus run them simultaneously. But I the configuration might take some time and might require installing some other application or patching ESD. I didn't try it. > But as we need something now and ESD seems not to be suitable, we probably have > to force people to replace ESD with something suitable. However I would not > require NAS, since nowadays there are many sondcards with hardware support for > multiple inputs on one DSP device, so we should also support direct DSP output. > I think there are two possibilities: > * using an interface library that is able at least of NAS and DSP output (maybe > the NAS library?). I think that the NAS library talks directly to /dev/dsp if NAS is not running. So it could be done this way. > * using direct DSP output and relying on audiooss. This is no win since I don't know about any suitable library for DSP other than the NAS library. Regards, Hynek From kirk@braille.uwo.ca Tue Jan 6 14:05:32 2004 From: Kirk Reiser To: speechd-discuss@nongnu.org Date: Sun Aug 13 12:52:45 2006 Subject: Speakup & Speech Dispatcher In-Reply-To: <20031226142123.GA1661@volny.cz> (message from Hynek Hanke on Fri, 26 Dec 2003 15:21:23 +0100) References: <3FD97DB6.5010601@brailcom.org> <20031226142123.GA1661@volny.cz> Message-ID: Well Hello Hynek and all. First of I apologize for the delay in getting back to your letter and Tomas's as well. I hope you all have had a very nice holiday season. I wouldn't mind hearing about some of your Xmas traditions at some time. I have really missed all of you over the past few months. I just wish I was better at correspondence than I tend to be. I also would like to be caught up on what you all have been doing since our trip back in July. Maybe we can have a speak freely conference or something. >I think we have to write some simple daemon application that would >read the commands from /dev/softsynth and then send them to Speech >Dispatcher in SSIP. It would act as a Speech Dispatcher client. This is kind of what I was hoping for. I have become more knowledgable about the various software synths since I wrote Tomas and have done a fair amount of work on Tuxtalk as well. The issue of getting the synth to shut-up as usual seems to be more of a problem than getting it to talk! 'wink' >This shouldn't be very difficult to write and I think I could do it >myself relatively soon. I didn't think is should be a very big job for anyone familiar with the ssip and speech dispatcher protocols. >I think maybe it would be easier and clearer to pass the commands >directly in SSIP. It would also make the task of writing the interface >daemon easier. Instead of creating your own command set. That is a possibility I chose the DoubleTalk command set because I am very comfortable with it not out of any conviction that it is a superior control language. >Also, I think that the Doubletalk protocol lacks some of the features >supported by SSIP, mainly the priority system. Software synthesis >is capable of doing more than the simple hardware devices and it would >be great if we could use the most of what current technology offers >(or just leave the doors open). I am not that fluent with the ssip protocol specification so I can't speak to the power or flexibility of one or the other. The DoubleTalk command set is quite extensive although I don't implement a lot of it. I mostly only include what is normal for basic communications with the synthesizers. I will happily outline the command set as I have implemented it if that would be useful. On the other hand I have no problem with modifying it to use ssip if it can be used in as simple a manner as what I have currently. We had some discussions about this interfacing back in the summer and as I indicated then, the kernel has no concept of priorities from an output perspective. So priorities would need to be imposed in an external daemon or client interface or somewhere. >Even if there is no direct use for priorities & history & sound icons >right now in Speakup, it might come later. You told me that you are >thinking about some application-specific customization of Speakup. When you >implement it, Speakup will be able to distinguish different messages >as less or more important and with the Doubletalk commands, you'll >have to invent yet another command set. That is definitely a future possibility. I haven't really given it much thought at this point so I can't say what we will want to do or can do at that point. Truthfully there are so many things which need to be done that what I'd like to do right now is get a good working solution for software synthesis and then put speakup to sleep for awhile so I can work on other projects such as tuxtalk, awe and javascript support for lynx. You are all quite aware of all the tasks which need attention. >Well, obviously, I don't know if there isn't any major obstacle to using >SSIP or it's derivative. Do you think it would be hard to replace the >current Doubetalk commands with SSIP commands or it's just a matter >of rewriting the strings here and there? I think outlining the two command sets will clarify how easy it would be to exchange ssip for the dtlk set. >I'm CC-ing this message to speechd@freebsoft.org. Maybe it would be >easier to communicate here. You are invited to join by sending >a mail to speechd-subscribe@freebsoft.org . I have subscribed and will continue the Cc:. Kirk From hanke@volny.cz Mon Jan 19 23:32:11 2004 From: Hynek Hanke To: speechd-discuss@nongnu.org Date: Sun Aug 13 12:52:45 2006 Subject: Speakup: Can Speakup distinguish terminals? Message-ID: <20040119223211.GB4496@volny.cz> Hello Kirk, I have one more question. Can you somehow distinguish in kernel what console are you reading the text from? If yes, is there a possibility to somehow pass the information to /dev/softsynth? This would be very useful, because the user could configure the interface daemon to use Speakup for some consoles while mute it for some others (where might be running speechd-el or some other specialized application with it's own speech interface). In other words, it might be desirable to switch, at run-time, for which consoles a user wants to use Speakup and for which not. This could be easily done in the daemon I'm writing right now, but only if I had some information about the origin of the different messages that are written to /dev/softsynth. I know the user can mute Speakup with pressing one or two keys, which is quite easy. But it might be annoying when frequently switching between consoles. It'd be better to be able to do it automatically. Do you think it would be possible? Regards, Hynek From hanke@volny.cz Mon Jan 19 23:19:58 2004 From: Hynek Hanke To: speechd-discuss@nongnu.org Date: Sun Aug 13 12:52:45 2006 Subject: Speakup & Speech Dispatcher (it talks!) In-Reply-To: References: <3FD97DB6.5010601@brailcom.org> <20031226142123.GA1661@volny.cz> Message-ID: <20040119221958.GA4496@volny.cz> On Tue, Jan 06, 2004 at 08:05:32AM -0500, Kirk Reiser wrote: > I will happily outline the command set as I have implemented it if > that would be useful. On the other hand I have no problem with > modifying it to use ssip if it can be used in as simple a manner as > what I have currently. Hi Kirk! Today I finally got to write some simple interface application from /dev/softsynth to speechd and it talks! :) I have yet to clean it a little bit and make it to be a true daemon. I don't think it is necessary for you to describe the protocol. I think it can be understood well from the code itself and the documentation to DoubleTalk. I've looked a little bit around the Speakup code and I think it isn't necessary to change the communication to SSIP now. The Doubletalk protocol serves it's purpose very well and switching to SSIP would most likely require some changes in the code. I have a few questions / notes about the two protocols: 1) TONE: I'm not very sure what does this mean. But if I understand it correctly from the docs, we don't have anything similar in SSIP. We prefer to set such voice properties in the synthesizer itself (here Festival or Flite), where it's more natural and where the user has better means to do such job. If one wants to use one voice with different "TONE" settings, he can create different virtual voices with such settings in Festival. This is quite different from hardware synthesis. So I don't know what to do with it? What is exactly the difference between TONE and PITCH? 2) VOLUME: Currently unsupported in Speech Dispatcher. The user can influence the volume through soundcard settings and a mixer. We will most likely add some volume settings in the future, but currently we have some problems to find a decent sound system/library. Temporarily marked as "yet unsupported". 3) FREQUENCY: I don't understand what this parameter does. Could you please explain? 4) What is the purpose of init_string, when is it sent to /dev/softsynth and what is the expected behaviour of the synthesizer? Can init_string be sent during the communication to reset the device or only at the beginning? I didn't notice any problems with the other commands. They seem to work right. Your Speakup module for software synthesis works perfectly! Thank you very much! Sincerely, Hynek From kirk@braille.uwo.ca Tue Jan 20 14:28:29 2004 From: Kirk Reiser To: speechd-discuss@nongnu.org Date: Sun Aug 13 12:52:45 2006 Subject: Speakup: Can Speakup distinguish terminals? In-Reply-To: <20040119223211.GB4496@volny.cz> (Hynek Hanke's message of "Mon, 19 Jan 2004 23:32:11 +0100") References: <20040119223211.GB4496@volny.cz> Message-ID: Hi Hynek: Let's see if I can remember all your questions first we'll do the terminal question proposed here. Speakup has no real notion of what console you are on other than a separate review structure is used for each console. So if you turn speakup off in console 4 let's say with the speakup-keypad-enter combination then the output of that console stays off until you specifically turn it back on. With that key combination you still have review functions so you can move around the screen reviewing what's there. If you use the prtscr (print screen) key speakup is totally turned off in the current console and there is no output review or otherwhise until the key is hit again to turn speakup back on. So you need only turn speakup off once in any console you wish to not have speech output in. You can also use the /proc/speakup system for various aspects of speakup including turning it on and off. Tone and frequency settings mean different things to different synth manufacturers. They are synonymous for most synths but in the case of the doubleTalk tone is the bigness or depth/quality of the voice while frequency is a two setting command which drops or raises the tone setting by a specific amount. Obviously not all synths support all commands so you basically just use the ones you have access to software synth or hardware synth and remove the structure members for command options your synth does not support or just ignore the sequences when intercepted. I have them included by default because I have written tuxtalk to support a number of the commands such as volume, tone, pitch, rate and punctuation levels. The init string is sent upon synth loading and then after the prtscr key is toggled back on to reset any settings which may have been lost while a hardware synth was turned off. In the case of most hardware synths the init string is also sent after a synth is turned back on after being off for a while with speakup still turned on. With hardware synths if the synth is turned off or removed speakup will notice after a few keys have been typed and turn it's output off so it needs to be re-initialized when the synth is connected once again. The key sequence speakup-kpad-enter checks to see if the synth has been turned off and then sends the init string if it can talk to the synth once again when issued. I'm not sure if I answered all your questions or not, if I missed any please re-ask. Kirk -- Kirk Reiser The Computer Braille Facility e-mail: kirk@braille.uwo.ca University of Western Ontario phone: (519) 661-3061 From pdm@freebsoft.org Wed Jan 21 10:23:08 2004 From: Milan Zamazal To: speechd-discuss@nongnu.org Date: Sun Aug 13 12:52:45 2006 Subject: Common Lisp SSIP interface available Message-ID: <87k73lfyf7.fsf@zamazal.org> If anyone is interested, Common Lisp interface to the Speech Synthesis Interface protocol is now available in CVS, see speechd/src/cl/. It was tested under CLisp and SBCL, porting to other systems should be easy. The interface may or may not be improved in future (e.g. properly documented), depending on your interest. Milan Zamazal -- The seeker after truth should be humbler than the dust. The world crushes the dust under its feet, but the seeker after truth should so humble himself that even the dust could crush him. -- M. K. Gandhi From hanke@volny.cz Sat Jan 31 00:29:52 2004 From: Hynek Hanke To: speechd-discuss@nongnu.org Date: Sun Aug 13 12:52:45 2006 Subject: Speakup works with Czech language Message-ID: <20040130232952.GA4264@volny.cz> Hi Kirk, how are you? I have some good news for you! We managed to get Speakup working with the Czech codepage. The trick is that in the international character sets, only some console fonts pass the keycodes properly to Speakup. So e.g. for the Czech language one must use the iso02.fxx.psf fonts (as oposed to lat2u-xx.psf or lat2-sunxx.psf. This isn't a difficult solution and also different character sets can be used on different consoles, which could be a good thing for the future. But I think it should be mentioned somewhere in the documentation, that only some console fonts work and that it's preferable to use the standard iso-* fonts, because although the other fonts also work with the iso character table, they somehow mess up the input to speakup. Now it talks Czech through Festival! Regards, Hynek From pdm@freebsoft.org Mon Feb 2 09:10:19 2004 From: Milan Zamazal To: speechd-discuss@nongnu.org Date: Sun Aug 13 12:52:45 2006 Subject: Speakup works with Czech language In-Reply-To: <20040130232952.GA4264@volny.cz> (Hynek Hanke's message of "Sat, 31 Jan 2004 00:29:52 +0100") References: <20040130232952.GA4264@volny.cz> Message-ID: <87znc17vh0.fsf@zamazal.org> >>>>> "HH" == Hynek Hanke writes: HH> The trick is that in the international character sets, only some HH> console fonts pass the keycodes properly to Speakup. HH> So e.g. for the Czech language one must use the iso02.fxx.psf HH> fonts (as oposed to lat2u-xx.psf or lat2-sunxx.psf. [...] HH> only some console fonts work and that it's preferable to use the HH> standard iso-* fonts, because although the other fonts also work HH> with the iso character table, they somehow mess up the input to HH> speakup. Could you elaborate about that, please? I know only very little about console fonts, so I wonder especially about the following things: - How are fonts related to keycodes? You mean that in the Linux kernel fonts and keyboards are somewhat related? - What's exactly wrong with the fonts? Do you have already got enough information to send appropriate bug reports? Regards, Milan Zamazal -- Wasting somebody else's time strikes me as the height of rudeness. Bill Gates From hanke@volny.cz Wed Feb 4 08:31:46 2004 From: Hynek Hanke To: speechd-discuss@nongnu.org Date: Sun Aug 13 12:52:45 2006 Subject: Speakup works with Czech language In-Reply-To: References: <20040130232952.GA4264@volny.cz> Message-ID: <20040204073146.GA1042@volny.cz> Hello Kirk, I've some news for you regarding Speakup internationalization and it's connection to Speech Dispatcher. I'll describe now what's necessary to make it work with Czech language in more detail: 1) speakup.c must be patched on one place. The spk_chartab table is specific to iso-8859-1 and it doesn't work with other codings (~l. 206, speakup.c), because in the extended character sets, some symbols over 128 are also ALPHA or A_CAP, not just B_SYM. If this is not changed, Speakup writes strange codes of these keys into /dev/softsynth (not the codes defined in iso-8859-2 for Czech, for example). Since it's not desirable to change Speakup's code for every internationalization and change it every time we want to use a different language, I'd like to ask if it wouldn't be possible to make this table accessible from /proc/speakup? So that users are able to change it like they are able to change /proc/speakup/characters. Then for example speechd-up would be able to set this table automatically, since there are routines in libc for telling if a given character, according to the given locales, is alpha, number or a capital alpha. 2) As for character reading and punctuation, /dec/proc/characters needs to be changed, of course. But we think it would be desirable if we could turn out these things in Speakup totally, so that we could let the software TTSs handle this task. There are good capabilities for processing punctuation and single characters in Festival, it's configurable for every language etc. So it would be best if Speakup just passes the "punctuation mode" DoubleTalk commands like it does now and don't process the text itself alone. If only one letter is passed, it's quite straightforward that it should be read as a single character. Again, we would welcome no processing on the Speakup's side. 3) Only some kinds of console screen fonts can be used for the Czech language. I don't really understand why, but it doesn't work with some console fonts, although all of them use iso-8859-2. If I use a different font than iso02.fxx.psf, the codes of the alphanumeric characters above 128 returned by Speakup don't correspond with iso-8859-2. It doesn't work with e.g. lat2u-xx.psf or lat2-sunxx.psf although these fonts are also iso-8859-2. I don't know what's the difference in them from the Speakup's perspective. But Mr. Buchal says it's the same with Brltty. The keycodes that brltty reads are also different according to the currently used console fonts. So I have a question: Where does Speakup actually read the input text from? Don't you know, how different fonts (although all working in the same character set) could influence this? If the above points are fulfilled, it works with Czech language. Something more might be needed for some other language, but I think this is sufficient for a large group of languages. We tested it with Speakup (cvs) loaded as a module, Speechd-up (cvs) interface daemon, Speech Dispatcher (cvs) and Festival 1.4.3. I've setup CVS for this interface daemon between Speakup and Speech Dispatcher which we called Speechd-Up to combine both names. Please look at http://www.freebsoft.org/speechd-up where there are information about the anonymous CVS acces and other things. After having downloaded the module, please see the file named INSTALL for more details about how to compile it. I'd like to ask some more questions: 1) How does CAPS_START and CAPS_STOP work and where and why is it used? It's not present in the documentation of the DoubleTalk protocol and I somehow don't understand it. 2) Wouldn't it be possible to pass also some command for setting language through /dev/softsynth? My idea is that the user could switch between two or three languages just by pressing a Speakup key shortcut. Then Speakup would send a command like ^XL2 or something like that to /dev/softsynth, Speechd-Up would look for what it's configured to do with language number 2, it would set the /proc/speakup/ tables accordingly, it would switch it's character set and synthesis language and the user could continue using this language. This way, software synthesis users would be able to easily switch languages. I think this partially applies also to hardware synthesis where this would be useful, but I'm not sure to what degree. 3) Speakup works fine! Thanks for your comments, Hynek From kirk@braille.uwo.ca Thu Feb 5 14:42:04 2004 From: Kirk Reiser To: speechd-discuss@nongnu.org Date: Sun Aug 13 12:52:45 2006 Subject: Speakup works with Czech language In-Reply-To: <20040204073146.GA1042@volny.cz> (Hynek Hanke's message of "Wed, 4 Feb 2004 08:31:46 +0100") References: <20040130232952.GA4264@volny.cz> <20040204073146.GA1042@volny.cz> Message-ID: Hello Hynek and all: If I have missed something please let me know. I have tried to answer your questions but how clearly those explanations will come accross is difficult to tell. I'll describe now what's necessary to make it work with Czech language in more detail: 1) speakup.c must be patched on one place. The spk_chartab table is specific to iso-8859-1 and it doesn't work with other codings (~l. 206, speakup.c), because in the extended character sets, some symbols over 128 are also ALPHA or A_CAP, not just B_SYM. If this is not changed, Speakup writes strange codes of these keys into /dev/softsynth (not the codes defined in iso-8859-2 for Czech, for example). Since it's not desirable to change Speakup's code for every internationalization and change it every time we want to use a different language, I'd like to ask if it wouldn't be possible to make this table accessible from /proc/speakup? So that users are able to change it like they are able to change /proc/speakup/characters. Then for example speechd-up would be able to set this table automatically, since there are routines in libc for telling if a given character, according to the given locales, is alpha, number or a capital alpha. I agree we need to figure out a way to conver the non-iso 1859-1 characters to get appropriate representations for various languages. The tables however which the kernel uses to represent those character sets are not part of speakup so I don't think we should make them available through the /proc/speakup interface. I am not exactly sure but it seems to me if the correct language tables are selected for the kernel and an appropriate translation were loaded into /proc/speakup/characters correct definitions would be sent to the synth soft or otherwhise. 2) As for character reading and punctuation, /dec/proc/characters needs to be changed, of course. But we think it would be desirable if we could turn out these things in Speakup totally, so that we could let the software TTSs handle this task. There are good capabilities for processing punctuation and single characters in Festival, it's configurable for every language etc. So it would be best if Speakup just passes the "punctuation mode" DoubleTalk commands like it does now and don't process the text itself alone. If only one letter is passed, it's quite straightforward that it should be read as a single character. Again, we would welcome no processing on the Speakup's side. I understand your point here and that's how we used to do it. We changed to using our own translation system because trying to support all the various synths we support meant for some synths we didn't have full representation. I implemented the /proc/speakup/character system so we could have a totally homogenious interface accross all synths and provide a mechanism so individuals could change those to suit their own needs. You see that of course because festival and flite use one way of pronouncing punctuation and the like while the Dectalk synth, tuxtalk and viavoice are different. The system is a simple character to string conversion so you can change to whatever you like for different synths. 3) Only some kinds of console screen fonts can be used for the Czech language. I don't really understand why, but it doesn't work with some console fonts, although all of them use iso-8859-2. If I use a different font than iso02.fxx.psf, the codes of the alphanumeric characters above 128 returned by Speakup don't correspond with iso-8859-2. It doesn't work with e.g. lat2u-xx.psf or lat2-sunxx.psf although these fonts are also iso-8859-2. I don't know what's the difference in them from the Speakup's perspective. There is no difference from a speakup perspective. You have to keep straight the difference between the language set tables and the font tables. The language set tables are single or double byte representations which get stored in video memory. One byte per character. The font tables are used to take those representations and turn them into bitmap image representations for the various video card character generators. So speakup reads the video memory to provide review functions and knows nothing about fonts or the like. So I have a question: Where does Speakup actually read the input text from? Don't you know, how different fonts (although all working in the same character set) could influence this? I believe I just answered this above if it still is not clear let me know and I'll try again. If the above points are fulfilled, it works with Czech language. Something more might be needed for some other language, but I think this is sufficient for a large group of languages. We tested it with Speakup (cvs) loaded as a module, Speechd-up (cvs) interface daemon, Speech Dispatcher (cvs) and Festival 1.4.3. I will check that out of cvs and play with it as soon as I finish the project I am currently working on. I'd like to ask some more questions: 1) How does CAPS_START and CAPS_STOP work and where and why is it used? It's not present in the documentation of the DoubleTalk protocol and I somehow don't understand it. Each synth has it's own command set definitions and capabilities as you know. Cap_start and cap_stop are speakups way of handling each synths nuances they are simple strings which get sent to the synth at the beginning and end of an uppercase letter or string of letters. That string can contain the control sequences for the synth or a text string like 'cap ' to be sent to the synth when any capital letter is encountered. These command set controls strings are contained in an array of pointers built in each synth driver to modify the output to be customizable for different synths. They are built at the bottom of each driver and a pointer to them is passed upon registration of the synth with the kernel. In the case of the Doubletalk the strings by default are ctrl-a+35p and ctrl-a-35p respectively to adjust the pitch up and then back down a noticible amount so when reviewing the screen a person knows immediately whether the character they are on is uppercase or not. 2) Wouldn't it be possible to pass also some command for setting language through /dev/softsynth? My idea is that the user could switch between two or three languages just by pressing a Speakup key shortcut. Then Speakup would send a command like ^XL2 or something like that to /dev/softsynth, Speechd-Up would look for what it's configured to do with language number 2, it would set the /proc/speakup/ tables accordingly, it would switch it's character set and synthesis language and the user could continue using this language. This way, software synthesis users would be able to easily switch languages. You could certainly do that. In fact synths like the Apollo that support multiple languages have such a sequence. In the doubletalk command set there is a command to change voices ctrl-anv to move between various voices. 'N' in this case is a number between 0-9 which could be used as an index for different voices. There is a /proc/speakup/voice entry created for synths that support multiple voices to give a simple way of changing between the voices. You certainly could set up keyboard macros to change languages/voices on the fly. One might extend the idea to desire changes in languages automagically when coming upon an umlaut or loading different applications. Those features come under the heading of automatic configuration sets which I have not implemented yet but it is on my agenda at some point. 3) Speakup works fine! Happy to hear that. I can hardly wait to try it!!!!. Kirk -- Kirk Reiser The Computer Braille Facility e-mail: kirk@braille.uwo.ca University of Western Ontario phone: (519) 661-3061 From hanke@volny.cz Fri Feb 6 18:28:37 2004 From: Hynek Hanke To: speechd-discuss@nongnu.org Date: Sun Aug 13 12:52:45 2006 Subject: Speakup works with Czech language In-Reply-To: References: <20040130232952.GA4264@volny.cz> <20040204073146.GA1042@volny.cz> Message-ID: <20040206172837.GA1563@volny.cz> On Thu, Feb 05, 2004 at 08:42:04AM -0500, Kirk Reiser wrote: >> 1) speakup.c must be patched on one place. The spk_chartab >> table is specific to iso-8859-1 and it doesn't work with >> other codings (~l. 206, speakup.c), because in the extended character >> sets, some symbols over 128 are also ALPHA or A_CAP, not just >> B_SYM. If this is not changed, Speakup writes strange codes >> of these keys into /dev/softsynth (not the codes defined >> in iso-8859-2 for Czech, for example). >> Since it's not desirable to change Speakup's code for every >> internationalization and change it every time we want to use a >> different language, I'd like to ask if it wouldn't be possible to make >> this table accessible from /proc/speakup? So that users are able to >> change it like they are able to change /proc/speakup/characters. > I agree we need to figure out a way to convert the non-iso 1859-1 > characters to get appropriate representations for various languages. We were discussing this here in Brailcom and we came to the conclusion that it would be the best solution to implement UTF-8 support into Speakup. There are two main reasons for this: 1) It solves the internationalization issue once forever. This way, people in all countries (except maybe for the ones who read right to left) will be able to use Speakup without any additional changes in the code. It solves the interaction with Speechd-Up since there will be no need to use different encodings. 2) The trend in text consoles in nations that don't use iso-8859-1 currently is to switch to Unicode. For example RedHat, by default, installs everything in UTF-8 since 9.0. If I understand it correctly, it isn't possible to run Speakup with UTF-8 now. It would be great if you could tell me what is needed for supporting UTF-8 in Speakup and what are the possible problems. You don't have to explain the basics of unicode, I'm using it quite extensively in Speech Dispatcher. But I'm not sure about the routines available in the kernel and so on. I don't know what the change would mean for the Speakup's code. Once we sort this out, I will be happy to help you with a part of this as I think it's very important. > The tables however which the kernel uses to represent those character > sets are not part of speakup so I don't think we should make them > available through the /proc/speakup interface. I was refering to speakup.c line 206 (the spk_chartab) array. This clearly is a part of Speakup and it doesn't work with other character sets than iso-8859-1. However, when we use unicode, this problem disappers. Since this is currently one of the most important things, I propose to leave the other things for now. I'll return to the question about punctuation processing later. Regards, Hynek From kirk@braille.uwo.ca Mon Feb 9 18:33:40 2004 From: Kirk Reiser To: speechd-discuss@nongnu.org Date: Sun Aug 13 12:52:45 2006 Subject: Speakup works with Czech language In-Reply-To: <20040206172837.GA1563@volny.cz> (Hynek Hanke's message of "Fri, 6 Feb 2004 18:28:37 +0100") References: <20040130232952.GA4264@volny.cz> <20040204073146.GA1042@volny.cz> <20040206172837.GA1563@volny.cz> Message-ID: Hi Hynek: I am sorry for the delay in getting back to you. I needed the time to look up the various things you were talking about. I apologize about the spk_chartab[] misunderstanding. I was thinking about the character array so I misunderstood what you were referring to. I would like to see that changed to reflect characters above 128 correctly. When that table was set-up it was simply done that way for expedience at the time. If there is a set of ranges which are common to most or all languages we should replace that section of the table with more appropriate values such as ALPHA or A_CAP so they'll be handled right no matter what character set is loaded. If you don't think there is a common set of values which would represent most char sets then we may have to look at another /proc/speakup file but I'd rather that be a last resort. If you have any ideas about that let me know. I will also look to see if I can find common sections of that range between various languages. As to your question about utf8 or unicode, speakup could handle that by modifying the call to speakup_con_write() in console.c line 2006 to pass the int instead of recasting to a char. The only thing I am reluctant about there is there someway of handling it in speakup.c to as not to need to define an array of 65536 elements because of the space requirement. Maybe we could define a two dimentional array with null lookups for sets which didn't exist or weren't necessary. It'd be very trivial to pass in the entire 16 bit argument though. It is after the conversion so the character is correct at that point and I strip off the high byte. I look forward to hearing your thoughts on this after reviewing the code. Kirk -- Kirk Reiser The Computer Braille Facility e-mail: kirk@braille.uwo.ca University of Western Ontario phone: (519) 661-3061 From kirk@braille.uwo.ca Mon Feb 9 19:53:33 2004 From: Kirk Reiser To: speechd-discuss@nongnu.org Date: Sun Aug 13 12:52:45 2006 Subject: Speakup works with Czech language In-Reply-To: <20040204073146.GA1042@volny.cz> (Hynek Hanke's message of "Wed, 4 Feb 2004 08:31:46 +0100") References: <20040130232952.GA4264@volny.cz> <20040204073146.GA1042@volny.cz> Message-ID: Hynek Hanke writes: > I've setup CVS for this interface daemon between Speakup and Speech > Dispatcher which we called Speechd-Up to combine both names. Please > look at > http://www.freebsoft.org/speechd-up Hi Again Hynek: I tried to go to this URL and variations on it with no luck. Could you please identify what I am doing wrong? Kirk -- Kirk Reiser The Computer Braille Facility e-mail: kirk@braille.uwo.ca University of Western Ontario phone: (519) 661-3061 From kirk@braille.uwo.ca Wed Feb 11 16:08:11 2004 From: Kirk Reiser To: speechd-discuss@nongnu.org Date: Sun Aug 13 12:52:45 2006 Subject: having various problems In-Reply-To: <87znc17vh0.fsf@zamazal.org> (Milan Zamazal's message of "Mon, 02 Feb 2004 09:10:19 +0100") References: <20040130232952.GA4264@volny.cz> <87znc17vh0.fsf@zamazal.org> Message-ID: Okay folks I have been running into various problems either installing speech dispatcher or compiling it. First of all Hynek, on the speechd-up web page I couldn't find any reference to the cvs module name so I had to download the tarball. I just thought you might want to know. On the other hand I tried installing the speech-dispatcher debian package on my machine at home and it quit with an error at the configuration stage because it couldn't find libflite_cmu-us... I don't remember the rest of the library name now but it looks to me that possibly the flite package doesn't install those libraries or there's a dependency problem with the package. I have checked out the speechd module on a machine here but it is having problems building the tools needed for compiling speech dispatcher. Firstly the INSTALL document doesn't mention needing to have libtool installed but my first attempt died because I didn't have it installed on that machine. The next set of errors I have a hunch are because there seems to be a problem with cvs checking out. I will include the output from an update I did this morning but I saw the same errors when doing the checkout yesterday but hadn't captured the output at that time. I will include the build.sh output below the cvs output but as I said I'm pretty sure it's a result of not having the entire tree to work with. Kirk ------------------------------------------------------------------------ ? Makefile.in ? aclocal.m4 ? autom4te.cache ? build.out ? config.h.in ? configure ? cvs.out ? make.out ? stamp-h.in ? config/Makefile.in ? config/clients/Makefile.in ? config/modules/Makefile.in ? doc/Makefile.in ? src/Makefile.in ? src/audio/Makefile.in ? src/c/Makefile.in ? src/c/api/Makefile.in ? src/c/clients/Makefile.in ? src/c/clients/say/Makefile.in cvs server: Updating . P ChangeLog cvs server: Updating config cvs server: Updating config/clients cvs server: Updating config/generic cvs server: Updating config/modules cvs server: Updating debian cvs server: Updating doc P doc/ChangeLog P doc/ssip.texi cvs server: Updating doc/figures cvs server: Updating doc/html cvs server: Updating intl P intl/fdset.h cvs server: Updating src cvs server: Updating src/audio P src/audio/spd_audio.c cvs server: Updating src/c cvs server: Updating src/c/api cvs server: Updating src/c/clients cvs server: Updating src/c/clients/say cvs server: Updating src/cl cvs server: failed to create lock directory for `/var/lib/cvs/speechd/src/cl' (/var/lock/cvs/speechd/src/cl/#cvs.lock): Permission denied cvs server: failed to obtain dir lock in repository `/var/lib/cvs/speechd/src/cl' cvs [server aborted]: read lock failed - giving up build.sh output ------------------------------------------------------------------------ Building user-defined autoconf macros (aclocal) Creating ./configure (autoconf) Creating config.h.in (autoheader) Checking for missing scripts (automake -a) automake: configure.in: installing `./install-sh' automake: configure.in: installing `./mkinstalldirs' automake: configure.in: installing `./missing' configure.in: 225: installing `./config.guess' src/Makefile.am:3: required directory src/server does not exist src/Makefile.am:3: required directory src/modules does not exist src/Makefile.am:3: required directory src/tests does not exist automake: src/c/clients/say/Makefile.am: not supported: source file `$(c_api)/libspeechd.h' is in subdirectory config/modules/Makefile.am:3: invalid variable `dist_moduleconf_DATA' config/clients/Makefile.am:5: invalid variable `dist_clientconf_DATA' doc/Makefile.am:2: installing `doc/mdate-sh' doc/Makefile.am:2: installing `doc/texinfo.tex' automake -a failed! -- Kirk Reiser The Computer Braille Facility e-mail: kirk@braille.uwo.ca University of Western Ontario phone: (519) 661-3061 From kirk@braille.uwo.ca Thu Feb 12 01:10:59 2004 From: Kirk Reiser To: speechd-discuss@nongnu.org Date: Sun Aug 13 12:52:45 2006 Subject: having various problems In-Reply-To: (Kirk Reiser's message of "Wed, 11 Feb 2004 10:08:11 -0500") References: <20040130232952.GA4264@volny.cz> <87znc17vh0.fsf@zamazal.org> Message-ID: Kirk Reiser writes: > On the other hand I tried installing the speech-dispatcher debian > package on my machine at home and it quit with an error at the > configuration stage because it couldn't find libflite_cmu-us... I > don't remember the rest of the library name now but it looks to me > that possibly the flite package doesn't install those libraries or > there's a dependency problem with the package. After playing around with the debian installation problem I found that it was a problem on my system. For some reason the libflite* libraries had been blown away and I have no idea how that happened. I apt-get --reinstalled the libflite1 package and all the missing libraries showed up and speech-dispatcher installed flawlessly. At last check though the anonymous cvs checkout is still broken. I'll try building speechd-up and see how it goes on the system with the debian installed packages. Kirk -- Kirk Reiser The Computer Braille Facility e-mail: kirk@braille.uwo.ca University of Western Ontario phone: (519) 661-3061 From hanke@volny.cz Wed Feb 11 23:52:53 2004 From: Hynek Hanke To: speechd-discuss@nongnu.org Date: Sun Aug 13 12:52:45 2006 Subject: having various problems In-Reply-To: References: <20040130232952.GA4264@volny.cz> <87znc17vh0.fsf@zamazal.org> Message-ID: <20040211225253.GD1391@volny.cz> On Wed, Feb 11, 2004 at 10:08:11AM -0500, Kirk Reiser wrote: > First of all Hynek, on the speechd-up web page I couldn't find any > reference to the cvs module name so I had to download the tarball. I > just thought you might want to know. This is a good point. This should be added here. The module name is "speechd-up". > I have checked out the speechd module on a machine here but it is having > problems building the tools needed for compiling speech dispatcher. > Firstly the INSTALL document doesn't mention needing to have libtool > installed It does. First, there is a list of general requirements. Then there is a list marked "Only if you are building from CVS, you also need these". > I will include the build.sh output below the cvs output but as I said > I'm pretty sure it's a result of not having the entire tree to work > with. Yes, it's mostly that. But it also seems you have an old version of automake, because it complains that some construct regarding subdirectories isn't supported. Automake 1.7 or newer is required. With Regards, Hynek From pdm@freebsoft.org Thu Feb 12 08:45:48 2004 From: Milan Zamazal To: speechd-discuss@nongnu.org Date: Sun Aug 13 12:52:45 2006 Subject: having various problems In-Reply-To: (Kirk Reiser's message of "Wed, 11 Feb 2004 19:10:59 -0500") References: <20040130232952.GA4264@volny.cz> <87znc17vh0.fsf@zamazal.org> Message-ID: <87u11w91w3.fsf@zamazal.org> >>>>> "KR" == Kirk Reiser writes: KR> At last check though the anonymous cvs checkout is still broken. It should be fixed now. KR> I'll try building speechd-up and see how it goes on the system KR> with the debian installed packages. If you experience any problems with the Debian packages, please report them through the Debian BTS. Thanks for pointing out the problems. Regards, Milan Zamazal -- All programmers are optimists. Perhaps this modern sorcery especially attracts those who believe in happy endings and fairy godmothers. Perhaps the hundreds of nitty frustrations drive away all but those who habitually focus on the end goal. -- Frederick P. Brooks, Jr. in The Mythical Man-Month From hanke@volny.cz Fri Feb 13 00:30:28 2004 From: Hynek Hanke To: speechd-discuss@nongnu.org Date: Sun Aug 13 12:52:46 2006 Subject: Speakup and UTF-8 In-Reply-To: References: <20040130232952.GA4264@volny.cz> <20040204073146.GA1042@volny.cz> <20040206172837.GA1563@volny.cz> Message-ID: <20040212233028.GA1872@volny.cz> On Mon, Feb 09, 2004 at 12:33:40PM -0500, Kirk Reiser wrote: > I apologize about the spk_chartab[] misunderstanding. I was thinking > about the character array so I misunderstood what you were referring > to. I would like to see that changed to reflect characters above 128 > correctly. Hi Kirk, I don't think there are common range across all the character sets. Maybe there are ranges about which we could say they represent alphanumeric characters, but definitely we can't distinguish small letters, capital letters and numbers implicitly. In some character sets, they are mixed together. But I really think there will be no need for this when Speakup uses UTF-8 internally. > As to your question about utf8 or unicode, speakup could handle that > by modifying the call to speakup_con_write() in console.c line 2006 to > pass the int instead of recasting to a char. Yes. I've checked this and it seems as a quite simple way how to input data to Speakup in full UTF-8. Then, all variables in Speakup that represent characters would have to be changed to int from char and arrays to int* instead of char*. But this shouldn't be hard for anyone who is familiar with the code. Also, the punctuation processing wouldn't work for characters above standard ascii, but I don't think this is a problem as I think this should be only a last option when the synthesizer doesn't support punctuation itself. It's definitely not an obstacle for Festival and other software synthesizers. > The only thing I am > reluctant about there is there someway of handling it in speakup.c to > as not to need to define an array of 65536 elements because of the > space requirement. I don't like this table at all. Why do you need it? There are some control characters and special characters in some range of the standard ascii. The others are just different letters and numbers. Why do you need to know, which letter is capital in Speakup? I think it would be better to pass the synthesizers just commands like "capital recognition on" and "capital recognition off". The information whether the letter is capital or not is carried in the letter itself. If there are synthesizers you want to support and they don't have capital recognition, it would be better to emulate it in their drivers instead. It should be fairly simple, but it will be restricted only to some parts of UTF-8 that you have tables for. But this doesn't matter, since the rest of the synthesizers will not be affected by this limitation. Speakup is a screen reader, I think it should only pass the appropriate text and set the right parameters on the synthesizers, not repeat the functionality of the synthesizers themselves. I'm definitely not against providing this option for the devices that don't have the capability of handling the more advanced things themselves, but let's not limit the better devices with that. If you take the thing as: ``all devices are the same, all of them are stupid'', then you will allways get only the ``higest common divisor''. I think Speakup could do better! So I propose to not distinguish between characters on a general level and maybe provide this capability for the devices that don't support this in their drivers (of course using some functions defined at one place, not repeating the code). You can consider if it's worth the effort (I know little about hardware devices). > Maybe we could define a two dimentional array with > null lookups for sets which didn't exist or weren't necessary. It'd be > very trivial to pass in the entire 16 bit argument though. It is > after the conversion so the character is correct at that point and I > strip off the high byte. I'm sorry, but I don't understand the last two sentences here very much. Could you please explain it in more detail? Have a nice day, Hynek From stivers_t@tomass.dyndns.org Fri Mar 12 17:31:32 2004 From: Thomas Stivers To: speechd-discuss@nongnu.org Date: Sun Aug 13 12:52:46 2006 Subject: question about speechd_up Message-ID: <20040312163132.GC4783@tomass.dyndns.org> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 First of all I would like to thank you for your part in getting software synthesizers to work with speakup. Now that I have it working one thing I have noticed is that sometimes the output from commands is not spoken and only the next prompt is spoken. This does not happen every time though. I suspect it may be an issue with the default priority speechd_up uses because if I read the spec correctly messages with priority text can interrupt themselves which is good for many things, but if one message is sent with part of a commands output and another is sent with the rest it would appear that the first message gets canceled. I just wanted to know if others had experienced this problem and had suggestions about how to fix it. Thanks in advance. - -- Clarke's Corollary: Any technology distinguishable from magic is insufficiently advanced. Thomas Stivers e-mail: stivers_t@tomass.dyndns.org -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.2.4 (GNU/Linux) iD8DBQFAUeXk5JK61UXLur0RAro+AJ9WC/oqrc4ZOAayO2aVa1e1dykYJwCeMcv1 JJWJlbABU/OpeA/KNYiZzlg= =QfoP -----END PGP SIGNATURE----- From jschmude@adelphia.net Fri Mar 12 22:55:50 2004 From: Jacob Schmude To: speechd-discuss@nongnu.org Date: Sun Aug 13 12:52:46 2006 Subject: speechd and software dectalk? Message-ID: <20040312215434.KKMU1437.mta10.adelphia.net@beavis> Hi everyone My name is Jacob Schmude and I'm a recent user of speechd. My primary use is to use speechd and speechd_up to communicate with the speakup screen reader. It works, but I'd like to get better speech quality. Flite just doesn't sound great, and festival isn't too responsive. Has anyone gotten speechd to communicate with software dectalk? I'm attempting to utilize the sd_generic driver to attempt to communicate with the dectalk's say program, but I'm not having much luck. I've got the command right, since when I enter the command on the command line it speaks, but when in speechd it doesn't work. The last message I get from the debug output is: speaking thread started... and that's it. Obviously, something's not right, since the debugging output for the flite module returns much more than that... and it doesn't speak. Has anyone successfully done this? If so, how? P.S. If this isn't the right place for this discussion, please point me in the right direction. From cerha@brailcom.org Sat Mar 13 12:42:21 2004 From: Tomas Cerha To: speechd-discuss@nongnu.org Date: Sun Aug 13 12:52:46 2006 Subject: speechd and software dectalk? In-Reply-To: <20040312215434.KKMU1437.mta10.adelphia.net@beavis> References: <20040312215434.KKMU1437.mta10.adelphia.net@beavis> Message-ID: <4052F39D.6040906@brailcom.org> Jacob Schmude napsal(a): > It works, but I'd like to get better speech quality. Flite just > doesn't sound great, and festival isn't too responsive. I remember we improved the responsivity by mounting /tmp to ramdisk, howerver I am not sure if this hack could still help these days. Could someone please make it clear? > Has anyone gotten speechd to communicate with software dectalk? Unfortunately not (as far as I know). Does the software dectalk have some API? It might not be difficult to write a true output driver for it. You can expect concrete aswers concerning the sd_generic driver from Hynek. Best regards, Tomas. -- Brailcom, o.p.s. http://www.brailcom.org Free(b)soft project http://www.freebsoft.org Eurochance project http://eurochance.brailcom.org From jschmude@adelphia.net Sat Mar 13 14:14:32 2004 From: Jacob Schmude To: speechd-discuss@nongnu.org Date: Sun Aug 13 12:52:46 2006 Subject: speechd and software dectalk? In-Reply-To: <4052F39D.6040906@brailcom.org> Message-ID: <20040313131417.VHYV2142.mta13.adelphia.net@beavis> Hi Yes, it does have an API. However, a programmer I am not, so I'm not sure how to implement it. Software dectalk is not free, either, so if someone else is ever going to try and attempt to build a driver for it, they'll have to shell out a bit of cash. On Sat, 13 Mar 2004 12:42:21 +0100, Tomas Cerha wrote: >Jacob Schmude napsal(a): >> It works, but I'd like to get better speech quality. Flite just >> doesn't sound great, and festival isn't too responsive. >I remember we improved the responsivity by mounting /tmp to ramdisk, howerver I >am not sure if this hack could still help these days. Could someone please >make it clear? >> Has anyone gotten speechd to communicate with software dectalk? >Unfortunately not (as far as I know). Does the software dectalk have some API? > It might not be difficult to write a true output driver for it. >You can expect concrete aswers concerning the sd_generic driver from Hynek. >Best regards, Tomas. >-- >Brailcom, o.p.s. http://www.brailcom.org >Free(b)soft project http://www.freebsoft.org >Eurochance project http://eurochance.brailcom.org -------------- next part -------------- An HTML attachment was scrubbed... URL: http://lists.freebsoft.org/pipermail/speechd/attachments/20040313/c012c983/attachment.html From hanke@volny.cz Sun Mar 14 00:37:54 2004 From: Hynek Hanke To: speechd-discuss@nongnu.org Date: Sun Aug 13 12:52:46 2006 Subject: speechd and software dectalk? In-Reply-To: <20040312215434.KKMU1437.mta10.adelphia.net@beavis> References: <20040312215434.KKMU1437.mta10.adelphia.net@beavis> Message-ID: <20040313233754.GC1531@volny.cz> On Fri, Mar 12, 2004 at 04:55:50PM -0500, Jacob Schmude wrote: > Has anyone gotten speechd to communicate with software dectalk? I'm attempting > to utilize the sd_generic driver to attempt to communicate with the dectalk's > say program, but I'm not having much luck. Hello Jacob, first of all, be prepared that the sd_generic output module is a highly sub-optimal solution. It might work well with the DecTalk's client but it doesn't have to. For example there is a requirement that the "client command" must stop speaking immediatelly when the process is killed. If you have some chance of doing it or getting someone to help you, it would be much better to write a native output module (as there is one for Flite and Festival) that would use DecTalk's API. But let's try. > The last message I get from the debug output is: > speaking thread started... It seems this output module didn't even get any requests to say something. So maybe your command line is ok, but I'm sure, it wasn't even executed. One possible cause is that you didn't set Speech Dispatcher in speechd.conf to use this output module as it's default module. If your module conf file is dectalk.conf, then you have to add: AddModule "dectalk" "sd_generic" "dectalk.conf" to speechd.conf and then modify the following lines in the same file: #LanguageDefaultModule "en" "flite" #LanguageDefaultModule "cs" "festival" #LanguageDefaultModule "es" "festival" (put the `#' sign before them to make them unactive) and DefaultModule dectalk to make it the default module. Then you have to restart Speech Dispatcher. If this doesn't help, please could you send me the module configuration file you are using for DecTalk in Speech Dispatcher? Also, please could you send me the logfile of a failure of this output module? I mean the one where you read the "speaking thread started" message. Please continue posting your emails to this group, it's the right place, but post these two files directly to me to hanke@volny.cz as they might be large and are not interesting for the rest of the group. > I've got the command right, since when I enter the command on the command > line it speaks, but when in speechd it doesn't work. Note that you can't type into the configuration file the command as you typed it in the terminal directly. Doublequotes `"' must be written as `\"' and if you split the command in more than one line, each of the lines except the last must be terminated by a backslash. > Has anyone successfully done this? If so, how? I've done a similar thing for the Epos synthesizer. You can look at epos-generic.conf for reference. But I don't know anything about DecTalk software. Good luck & thanks for your message, Hynek Hanke From hanke@volny.cz Sun Mar 14 00:04:17 2004 From: Hynek Hanke To: speechd-discuss@nongnu.org Date: Sun Aug 13 12:52:46 2006 Subject: question about speechd_up In-Reply-To: <20040312163132.GC4783@tomass.dyndns.org> References: <20040312163132.GC4783@tomass.dyndns.org> Message-ID: <20040313230417.GB1531@volny.cz> On Fri, Mar 12, 2004 at 10:31:32AM -0600, Thomas Stivers wrote: > First of all I would like to thank you for your part in getting software > synthesizers to work with speakup. Now that I have it working one thing > I have noticed is that sometimes the output from commands is not spoken > and only the next prompt is spoken. This does not happen every time > though. I suspect it may be an issue with the default priority > speechd_up uses because if I read the spec correctly messages with > priority text can interrupt themselves which is good for many things, > but if one message is sent with part of a commands output and another is > sent with the rest it would appear that the first message gets canceled. Hello Thomas, when the sftsyn_speakup driver writes it's commands to /dev/softsynth, it usually sends a STOP command before each message. So I thought that it's ok to send the messages with priority TEXT to Speech Dispatcher, which does exactly the same thing, but a little bit faster since the stops are comming from speakup only after the message. But maybe I was mistaken and there are cases where a message from Speakup isn't preceeded by a STOP command. If this is the case, we can send the messages to Speech Dispatcher with a different priority (MESSAGE) which is not interrupted by newly arriving messages and leave the interrupting of speech solely on the explicit STOP commands from /dev/softsynth. It seems you know much better how should Speakup really work (I've no reference here, my only device is software synthesis). Please could you, on line 227 and 282 of speechd-up.c, rewrite the priority SPD_TEXT to SPD_MESSAGE, both lines should read spd_say(conn, SPD_MESSAGE, text); and tell us if this fixes the problem? Thank you for your report! Hynek From hanke@volny.cz Sun Mar 14 18:02:12 2004 From: Hynek Hanke To: speechd-discuss@nongnu.org Date: Sun Aug 13 12:52:46 2006 Subject: speechd and software dectalk? In-Reply-To: <20040312215434.KKMU1437.mta10.adelphia.net@beavis> References: <20040312215434.KKMU1437.mta10.adelphia.net@beavis> Message-ID: <20040314170212.GI1020@volny.cz> On Fri, Mar 12, 2004 at 04:55:50PM -0500, Jacob Schmude wrote: > Has anyone gotten speechd to communicate with software dectalk? > I'm attempting to utilize the sd_generic driver to attempt to communicate > with the dectalk's say program, but I'm not having much luck. I'm writing just to let others know that with Jacob's help, we finally managed to get the generic driver working with the DTK software synthesizer client. If you are interested, it's now included in the CVS version of speechd. But please be aware that DTK software synthesizer is not Free Software. Regards, Hynek Hanke From pdm@freebsoft.org Mon Mar 15 12:04:50 2004 From: Milan Zamazal To: speechd-discuss@nongnu.org Date: Sun Aug 13 12:52:46 2006 Subject: speechd and software dectalk? In-Reply-To: <20040312215434.KKMU1437.mta10.adelphia.net@beavis> (Jacob Schmude's message of "Fri, 12 Mar 2004 16:55:50 -0500") References: <20040312215434.KKMU1437.mta10.adelphia.net@beavis> Message-ID: <87k71m8j71.fsf@zamazal.org> >>>>> "JS" == Jacob Schmude writes: JS> festival isn't too responsive. What does it mean exactly? Perhaps it's a problem that might be fixed. >>>>> "TC" == Tomas Cerha writes: TC> I remember we improved the responsivity by mounting /tmp to TC> ramdisk, howerver I am not sure if this hack could still help TC> these days. That was a problem of a particular computer, irreproducible anywhere else. Milan Zamazal -- _/_\_/_ o _\_/_\_ o _/_\_/_ o _\_/_\_ o BEWARE! -<_|_|_|_><-- -<_|_|_|_><-- -<_|_|_|_><-- -<_|_|_|_><-- *Bugs* are / \ / o \ / \ o / \ / o \ / \ o approaching! From pdm@freebsoft.org Tue Mar 30 11:50:30 2004 From: Milan Zamazal To: speechd-discuss@nongnu.org Date: Sun Aug 13 12:52:46 2006 Subject: Announce: speechd-el 0.3 released Message-ID: <871xnak6jd.fsf@zamazal.org> speechd-el 0.3 released ======================= The Free(b)soft project is pleased to announce speechd-el 0.3. * What is speechd-el? speechd-el is an Emacs client to Speech Dispatcher, providing a complex speech interface to Emacs, focused especially on (but not limited to) the blind and visually impaired users. It allows the user to work with Emacs without looking on the screen, using the speech output produced by the synthesizers supported in Speech Dispatcher. Key speechd-el features are: - Automated speaking of most Emacs messages and actions, without necessity to customize each particular Emacs function to make it speak. - Most built-in and external Emacs packages can produce speech output immediately, without any special support. - Almost no change of the standard Emacs behavior, no interference with user settings. - Highly configurable. - Full cooperation with Speech Dispatcher. - Small code size. - Free Software, distributed under the GPL. Part of the speechd-el package is an Elisp library for communication with Speech Dispatcher, allowing you to write your own Emacs speech output extensions and your own Emacs interfaces to Speech Dispatcher. * What is new in the 0.3 version? - Connection parameters and voice handling configuration was simplified and made more flexible. - The *Completion* buffer is spoken now. - Volume setting added. - You can shut up speechd-el in Elisp debuggers. - Bug reporting improvements. - Bug fixes. See the NEWS file for more details. * Where to get it? You can get the distribution tarball of the released version from http://www.freebsoft.org/pub/projects/speechd-el/speechd-el-0.3.tar.gz . The home page of the project is http://www.freebsoft.org/speechd-el . From jon@squinky.gotdns.com Wed Mar 31 00:58:20 2004 From: Jon Hood To: speechd-discuss@nongnu.org Date: Sun Aug 13 12:52:46 2006 Subject: speechd broken pipe to module Message-ID: <1080687500.6667.15.camel@localhost> Below is a problem i have been having with speech-dispatcher. Please let me know if you see the problem. Also, would it be very easy to move the config files to /etc/speech-dispatcher/ instead of /etc/speechd/? How would I go about doing that? -Jon commands executed: # /etc/init.d/speech-dispatcher start * Starting festival... [ ok ] * Starting speech-dispatcher...[ ok ] # say "test" //no sound plays (festival) # cat /var/log/speech-dispatcher.log //sorry about its length [Tue Mar 30 16:50:24 2004] speechd: Speech Dispatcher Logging to file /var/log/speech-dispatcher.log [Tue Mar 30 16:50:24 2004] speechd: Initializing output module flite with binary /usr/lib/speech-dispatcher-modules/sd_flite and configuration /etc/speechd/modules//flite.conf [Tue Mar 30 16:50:24 2004] speechd: Module flite loaded. [Tue Mar 30 16:50:24 2004] speechd: Module name=flite being inserted into hash table [Tue Mar 30 16:50:24 2004] speechd: Initializing output module festival with binary /usr/lib/speech-dispatcher-modules/sd_festival and config uration /etc/speechd/modules//festival.conf [Tue Mar 30 16:50:24 2004] speechd: Module festival loaded. [Tue Mar 30 16:50:24 2004] speechd: Module name=festival being inserted into hash table [Tue Mar 30 16:50:24 2004] speechd: Configuration has been read from "/etc/speech-dispatcher/speechd.conf" [Tue Mar 30 16:50:24 2004] speechd: Speech Dispatcher started with 2 output modules [Tue Mar 30 16:50:24 2004] speechd: Creating new thread for speak() [Tue Mar 30 16:50:24 2004] speechd: Semaphore waiting [Tue Mar 30 16:50:24 2004] speechd: Semaphore returned [Tue Mar 30 16:50:24 2004] speechd: Semaphore waiting [Tue Mar 30 16:50:24 2004] speechd: Openning a socket connection [Tue Mar 30 16:50:24 2004] speechd: Speech Dispatcher waiting for clients ... [Tue Mar 30 16:50:57 2004] speechd: Activity on fd 12 ... [Tue Mar 30 16:50:57 2004] speechd: Adding client on fd 13 [Tue Mar 30 16:50:57 2004] speechd: Data structures for client on fd 13 created [Tue Mar 30 16:50:57 2004] speechd: Activity on fd 13 ... [Tue Mar 30 16:50:57 2004] speechd: Read 38 bytes from client on fd 13 [Tue Mar 30 16:50:57 2004] speechd: 13:DATA:|SET SELF CLIENT_NAME "root:say:main" | [Tue Mar 30 16:50:57 2004] speechd: Command caught: "set" [Tue Mar 30 16:50:57 2004] speechd: 13:REPLY:|208 OK CLIENT NAME SET | [Tue Mar 30 16:50:57 2004] speechd: Activity on fd 13 ... [Tue Mar 30 16:50:57 2004] speechd: Read 24 bytes from client on fd 13 [Tue Mar 30 16:50:57 2004] speechd: 13:DATA:|SET SELF PRIORITY TEXT | [Tue Mar 30 16:50:57 2004] speechd: Command caught: "set" [Tue Mar 30 16:50:57 2004] speechd: 13:REPLY:|202 OK PRIORITY SET | [Tue Mar 30 16:50:57 2004] speechd: Activity on fd 13 ... [Tue Mar 30 16:50:57 2004] speechd: Read 7 bytes from client on fd 13 [Tue Mar 30 16:50:57 2004] speechd: 13:DATA:|SPEAK | [Tue Mar 30 16:50:57 2004] speechd: Command caught: "speak" [Tue Mar 30 16:50:57 2004] speechd: Switching to data mode... [Tue Mar 30 16:50:57 2004] speechd: 13:REPLY:|230 OK RECEIVING DATA | [Tue Mar 30 16:50:57 2004] speechd: Activity on fd 13 ... [Tue Mar 30 16:50:57 2004] speechd: Read 4 bytes from client on fd 13 [Tue Mar 30 16:50:57 2004] speechd: 13:DATA:|test| [Tue Mar 30 16:50:57 2004] speechd: Buffer: |test| bytes: [Tue Mar 30 16:50:57 2004] speechd: Activity on fd 13 ... [Tue Mar 30 16:50:57 2004] speechd: Read 5 bytes from client on fd 13 [Tue Mar 30 16:50:57 2004] speechd: 13:DATA:| . | [Tue Mar 30 16:50:57 2004] speechd: Buffer: | . | bytes: [Tue Mar 30 16:50:57 2004] speechd: Finishing data [Tue Mar 30 16:50:57 2004] speechd: Switching back to command mode... [Tue Mar 30 16:50:57 2004] speechd: Incomming text: |test| [Tue Mar 30 16:50:57 2004] speechd: Altering text (I): || [Tue Mar 30 16:50:57 2004] speechd: Altering text (II): || [Tue Mar 30 16:50:57 2004] speechd: Altered text: |test| [Tue Mar 30 16:50:57 2004] speechd: New buf is now: |test| [Tue Mar 30 16:50:57 2004] speechd: In queue_message desired output module is festival [Tue Mar 30 16:50:57 2004] speechd: Queueing message |test| with priority 2 [Tue Mar 30 16:50:57 2004] speechd: Semaphore post [Tue Mar 30 16:50:57 2004] speechd: Posting on semaphore. [Tue Mar 30 16:50:57 2004] speechd: 13:REPLY:|225 OK MESSAGE QUEUED | [Tue Mar 30 16:50:57 2004] speechd: Semaphore returned [Tue Mar 30 16:50:57 2004] speechd: Message in queue p3 [Tue Mar 30 16:50:57 2004] speechd: MSG before index marking: |test| [Tue Mar 30 16:50:57 2004] speechd: MSG after index marking: |test| [Tue Mar 30 16:50:57 2004] speechd: Desired output module is festival [Tue Mar 30 16:50:57 2004] speechd: Incomming text: |test| [Tue Mar 30 16:50:57 2004] speechd: Altering text (I): || [Tue Mar 30 16:50:57 2004] speechd: Altering text (II): || [Tue Mar 30 16:50:57 2004] speechd: Altered text: |test| [Tue Mar 30 16:50:57 2004] speechd: Module set. [Tue Mar 30 16:50:57 2004] speechd: Command sent to output module: |SET | [Tue Mar 30 16:50:57 2004] speechd: Activity on fd 13 ... [Tue Mar 30 16:50:57 2004] speechd: Read 6 bytes from client on fd 13 [Tue Mar 30 16:50:57 2004] speechd: 13:DATA:|QUIT | [Tue Mar 30 16:50:57 2004] speechd: Command caught: "quit" [Tue Mar 30 16:50:57 2004] speechd: Bye received. [Tue Mar 30 16:50:57 2004] speechd: Removing client on fd 13 [Tue Mar 30 16:50:57 2004] speechd: Tagging client as inactive in settings [Tue Mar 30 16:50:57 2004] speechd: Removing client from the fd->uid table. [Tue Mar 30 16:50:57 2004] speechd: Closing clients file descriptor 13 [Tue Mar 30 16:50:57 2004] speechd: Connection closed [Tue Mar 30 16:50:57 2004] speechd: Reply from output module: |203 OK RECEIVING SETTINGS | [Tue Mar 30 16:50:57 2004] speechd: Command sent to output module: |pitch=0 rate=0 punctuation_mode=none spelling_mode=off cap_let_recogn=none language=en voice=male1 | [Tue Mar 30 16:50:57 2004] speechd: Command sent to output module: |. | [Tue Mar 30 16:50:57 2004] speechd: Reply from output module: |203 OK SETTINGS RECEIVED | [Tue Mar 30 16:50:57 2004] speechd: Module speak! [Tue Mar 30 16:50:57 2004] speechd: Command sent to output module: |SPEAK | [Tue Mar 30 16:50:57 2004] speechd: Reply from output module: |202 OK RECEIVING MESSAGE | [Tue Mar 30 16:50:57 2004] speechd: Command sent to output module: |test| [Tue Mar 30 16:50:57 2004] speechd: Command sent to output module: | . | [Tue Mar 30 16:50:57 2004] speechd: Broken pipe to module. [Tue Mar 30 16:50:57 2004] speechd: Output module working status: 0 (pid:7283) [Tue Mar 30 16:50:57 2004] speechd: Output module terminated abnormally, probably crashed. [Tue Mar 30 16:50:57 2004] speechd: Message sent to output module [Tue Mar 30 16:50:57 2004] speechd: Error: Output module failed [Tue Mar 30 16:50:57 2004] speechd: Output module working status: 0 (pid:0) [Tue Mar 30 16:50:57 2004] speechd: Output module terminated abnormally, probably crashed. [Tue Mar 30 16:50:57 2004] speechd: Semaphore waiting -- AIM: Squinky01 Yahoo!: Squinky86 ICQ: 160940989 -------------- next part -------------- A non-text attachment was scrubbed... Name: not available Type: application/pgp-signature Size: 189 bytes Desc: This is a digitally signed message part Url : http://lists.freebsoft.org/pipermail/speechd/attachments/20040330/b386305c/attachment.pgp From hanke@volny.cz Wed Mar 31 01:52:14 2004 From: Hynek Hanke To: speechd-discuss@nongnu.org Date: Sun Aug 13 12:52:46 2006 Subject: speechd broken pipe to module In-Reply-To: <1080687500.6667.15.camel@localhost> References: <1080687500.6667.15.camel@localhost> Message-ID: <20040330235214.GA20316@volny.cz> On Tue, Mar 30, 2004 at 04:58:20PM -0600, Jon Hood wrote: > Below is a problem i have been having with speech-dispatcher. Please let > me know if you see the problem. Hi Jon, thank you for your report. It seems there is something wrong happening in the Festival output module, probably a bug. But it can't be seen exactly from this log you sent because the output modules are separate processes. So, please could you turn the debugging in the output module on (see /etc/speechd/modules/festival.conf), Debug 1 DebugFile "/tmp/debug-festival" duplicate the error and then send me the output (/tmp/debug-festival/)? This would greatly help me in finding the critical place where something goes wrong. Thank you very much. > Also, would it be very easy to move the > config files to /etc/speech-dispatcher/ instead of /etc/speechd/? How > would I go about doing that? Yes, you can edit configure.in (the spdconfdir and moduleconfdir variables) in the source directory and then recompile the package. Also, you will have to change the default path for Include files in speechd.conf (you only need this if you include some other configuration files in the main file). But if you wait a little bit, we are near making a new release where the speech-dispatcher directories are used by default, exactly as you describe. If things go well, it might be available at the end of this week. Or you can download the CVS version in the meantime. Regards, Hynek Hanke From jon@squinky.gotdns.com Sun Apr 4 21:49:16 2004 From: Jon Hood To: speechd-discuss@nongnu.org Date: Sun Aug 13 12:52:46 2006 Subject: speechd broken pipe to module In-Reply-To: <20040330235214.GA20316@volny.cz> References: <1080687500.6667.15.camel@localhost> <20040330235214.GA20316@volny.cz> Message-ID: <1081108156.16012.5.camel@localhost> On Tue, 2004-03-30 at 17:52, Hynek Hanke wrote: > So, please could you turn the debugging in the output module on > (see /etc/speechd/modules/festival.conf), > Debug 1 > DebugFile "/tmp/debug-festival" duplicate the error and then > send me the output (/tmp/debug-festival/)? This would greatly help me > in finding the critical place where something goes wrong. Sorry it's taking so long for me to get done with all this, I've been playing around with the cvs version for a while, and I realize that it doesn't matter what module I chose, I get the broken pipe to module error. Note that if I run: $ echo '(SayText "hi")' | telnet localhost 1314 festival will say "hi" just fine. Below is /tmp/debug-module (note that there is some unknown character after "Festival: leaving write() normaly"): # spd-say test # cat /tmp/debug-module Sun Apr 4 14:43:33 2004 [688531]: module_init() Sun Apr 4 14:43:33 2004 [693661]: -> Festival: |(require 'speech-dispatcher) | Sun Apr 4 14:43:33 2004 [712048]: <- Festival: |LP | Sun Apr 4 14:43:33 2004 [712105]: <- Festival: |OK | Sun Apr 4 14:43:33 2004 [712157]: -> Festival: |(Parameter.set 'Wavefiletype 'nist) | Sun Apr 4 14:43:33 2004 [712745]: <- Festival: |LP | Sun Apr 4 14:43:33 2004 [752238]: <- Festival: |OK | Sun Apr 4 14:43:33 2004 [752299]: FestivalServerHost = localhost Sun Apr 4 14:43:33 2004 [752316]: FestivalServerPort = 1314 Sun Apr 4 14:43:33 2004 [752396]: Festival: creating new thread for festival_speak Sun Apr 4 14:43:33 2004 [752646]: festival: speaking thread starting....... Sun Apr 4 14:43:33 2004 [752685]: Cache: initialized Sun Apr 4 14:43:40 2004 [82735]: module_write() Sun Apr 4 14:43:40 2004 [82866]: -> Festival: |(speechd-set-language "en") | Sun Apr 4 14:43:40 2004 [84667]: <- Festival: |LP | Sun Apr 4 14:43:40 2004 [84758]: <- Festival: |OK | Sun Apr 4 14:43:40 2004 [84824]: -> Festival: |(speechd-set-voice "male1") | Sun Apr 4 14:43:40 2004 [85907]: <- Festival: |LP | Sun Apr 4 14:43:40 2004 [125199]: <- Festival: |OK | Sun Apr 4 14:43:40 2004 [125276]: Filling new sample wave Sun Apr 4 14:43:40 2004 [125355]: -> Festival: |(speechd-set-rate 0) | Sun Apr 4 14:43:40 2004 [126150]: <- Festival: |LP | Sun Apr 4 14:43:40 2004 [166178]: <- Festival: |OK | Sun Apr 4 14:43:40 2004 [166295]: -> Festival: |(speechd-set-pitch 0) | Sun Apr 4 14:43:40 2004 [167144]: <- Festival: |LP | Sun Apr 4 14:43:40 2004 [207162]: <- Festival: |OK | Sun Apr 4 14:43:40 2004 [207300]: -> Festival: |(speechd-set-punctuation-mode 'none) | Sun Apr 4 14:43:40 2004 [208118]: <- Festival: |LP | Sun Apr 4 14:43:40 2004 [248130]: <- Festival: |OK | Sun Apr 4 14:43:40 2004 [248248]: -> Festival: |(speechd-set-capital-character-recognition-mode nil) | Sun Apr 4 14:43:40 2004 [248937]: <- Festival: |LP | Sun Apr 4 14:43:40 2004 [288140]: <- Festival: |OK | Sun Apr 4 14:43:40 2004 [288197]: Requested data: |test| (recoding to ISO-8859-1 ) Sun Apr 4 14:43:40 2004 [288699]: Requested data after recoding: |test| Sun Apr 4 14:43:40 2004 [288737]: Requested after stripping punct: |test| Sun Apr 4 14:43:40 2004 [288775]: Festival: leaving write() normaly Sun Apr 4 14:43:40 2004 [288860]: Semaphore on -- AIM: Squinky01 Yahoo!: Squinky86 ICQ: 160940989 -------------- next part -------------- A non-text attachment was scrubbed... Name: not available Type: application/pgp-signature Size: 189 bytes Desc: This is a digitally signed message part Url : http://lists.freebsoft.org/pipermail/speechd/attachments/20040404/5603b551/attachment.pgp From jon@squinky.gotdns.com Sun Apr 4 21:52:53 2004 From: Jon Hood To: speechd-discuss@nongnu.org Date: Sun Aug 13 12:52:46 2006 Subject: speechd broken pipe to module In-Reply-To: <1081108156.16012.5.camel@localhost> References: <1080687500.6667.15.camel@localhost> <20040330235214.GA20316@volny.cz> <1081108156.16012.5.camel@localhost> Message-ID: <1081108373.16017.10.camel@localhost> forgot to mention that I'm using a 2.6.3 kernel, if that's any help -------------- next part -------------- A non-text attachment was scrubbed... Name: not available Type: application/pgp-signature Size: 189 bytes Desc: This is a digitally signed message part Url : http://lists.freebsoft.org/pipermail/speechd/attachments/20040404/c044d0ec/attachment.pgp From hanke@volny.cz Mon Apr 12 21:24:31 2004 From: Hynek Hanke To: speechd-discuss@nongnu.org Date: Sun Aug 13 12:52:46 2006 Subject: speechd broken pipe to module In-Reply-To: <1081108156.16012.5.camel@localhost> References: <1080687500.6667.15.camel@localhost> <20040330235214.GA20316@volny.cz> <1081108156.16012.5.camel@localhost> Message-ID: <20040412192431.GK1207@volny.cz> On Sun, Apr 04, 2004 at 02:49:16PM -0500, Jon Hood wrote: > Sorry it's taking so long for me to get done with all this, I've been > playing around with the cvs version for a while, and I realize that it > doesn't matter what module I chose, I get the broken pipe to module > error. After discussing this privately with Jon, we found that the problem was probably fixed in the 0.3 release. With Regards, Hynek Hanke From jon@squinky.gotdns.com Mon Apr 12 16:41:26 2004 From: Jon Hood To: speechd-discuss@nongnu.org Date: Sun Aug 13 12:52:46 2006 Subject: festival voices Message-ID: <1081780885.6881.7.camel@localhost> Thanks for helping me with the other issue, Hynek :). Now that festival is working with speech-dispatcher, I get to bother everyone with other issues! Right now, I'm wondering how to get different voices in festival. So far I have the following lines added to modules/festival.conf: AddVoice "en" "MALE1" "kal_diphone" AddVoice "en" "MALE2" "ked_diphone" AddVoice "en" "MALE3" "don_diphone" AddVoice "en" "FEMALE1" "us1_mbrola" festival> (voice.list) (us2_mbrola rab_diphone ked_diphone don_diphone us1_mbrola us3_mbrola kal_diphone el_diphone) The mbrola voices are set up correctly, but when I put FEMALE1 in speechd.conf, I do not get the us1_mbrola voice in festival. I can tell that MALE1 and MALE2 are different, but MALE3 and FEMALE1 sound the same as MALE1. -Jon -- AIM: Squinky01 Yahoo!: Squinky86 ICQ: 160940989 Jabber: squinky86@im.gentoo.org -------------- next part -------------- A non-text attachment was scrubbed... Name: not available Type: application/pgp-signature Size: 189 bytes Desc: This is a digitally signed message part Url : http://lists.freebsoft.org/pipermail/speechd/attachments/20040412/627efb32/attachment.pgp From pdm@freebsoft.org Tue Apr 13 08:05:18 2004 From: Milan Zamazal To: speechd-discuss@nongnu.org Date: Sun Aug 13 12:52:46 2006 Subject: festival voices In-Reply-To: <1081780885.6881.7.camel@localhost> (Jon Hood's message of "Mon, 12 Apr 2004 14:41:26 +0000") References: <1081780885.6881.7.camel@localhost> Message-ID: <878yh0jtup.fsf@zamazal.org> >>>>> "JH" == Jon Hood writes: JH> Right now, I'm wondering how to get different voices in JH> festival. So far I have the following lines added to JH> modules/festival.conf: Setting voices in that file is obsolete and doesn't work anymore. Instead, set them in your Festival init file (e.g. /etc/festival.scm), variable speechd-language-voices. The default value of the variable is set with (set! speechd-language-voices '((english (male1 voice_kal_diphone) (male2 voice_ked_diphone)) (americanenglish (male1 voice_kal_diphone) (male2 voice_ked_diphone)) (britishenglish (male1 voice_kal_diphone) (male2 voice_ked_diphone)) (german (male1 voice_german)) (czech (male1 voice_czech_mbrola_cz2))) I believe it's obvious how to add more voices. Regards, Milan Zamazal -- http://www.zamazal.org From kirk@braille.uwo.ca Wed Apr 14 14:03:50 2004 From: Kirk Reiser To: speechd-discuss@nongnu.org Date: Sun Aug 13 12:52:46 2006 Subject: changes to speechd-up Message-ID: Hello all, In responce to some problems that Jacob has had with the speechd-up module and speakup I have made some changes to the module. I wondered whether you prefer I send diff's here to the list or if you wanted to give me write permission to the cvs repository. Currently the changes are quite small but I am hoping to expand a bit on the command set code. I believe that Jacob has added a bit of code to that as well. Kirk -- Kirk Reiser The Computer Braille Facility e-mail: kirk@braille.uwo.ca University of Western Ontario phone: (519) 661-3061 From jon@squinky.gotdns.com Thu Apr 15 18:13:25 2004 From: Jon Hood To: speechd-discuss@nongnu.org Date: Sun Aug 13 12:52:46 2006 Subject: festival voices In-Reply-To: <878yh0jtup.fsf@zamazal.org> References: <1081780885.6881.7.camel@localhost> <878yh0jtup.fsf@zamazal.org> Message-ID: <1082045604.6787.8.camel@localhost> On Tue, 2004-04-13 at 06:05, Milan Zamazal wrote: > (set! speechd-language-voices > '((english > (male1 voice_kal_diphone) > (male2 voice_ked_diphone)) > (americanenglish > (male1 voice_kal_diphone) > (male2 voice_ked_diphone)) > (britishenglish > (male1 voice_kal_diphone) > (male2 voice_ked_diphone)) > (german > (male1 voice_german)) > (czech > (male1 voice_czech_mbrola_cz2))) I added the following to /usr/lib/festival/init.scm and set "female1" in speechd.conf, but still no beans: (set! speechd-language-voices '((english (male1 voice_kal_diphone) (male2 voice_ked_diphone) (female1 voice_us1_mbrola)) (americanenglish (male1 voice_kal_diphone) (male2 voice_ked_diphone)) (britishenglish (male1 voice_kal_diphone) (male2 voice_ked_diphone)) (german (male1 voice_german)) (czech (male1 voice_czech_mbrola_cz2))) ) I am using speech-dispatcher-0.3. -Jon -- AIM: Squinky01 Yahoo!: Squinky86 ICQ: 160940989 Jabber: squinky86@im.gentoo.org -------------- next part -------------- A non-text attachment was scrubbed... Name: not available Type: application/pgp-signature Size: 189 bytes Desc: This is a digitally signed message part Url : http://lists.freebsoft.org/pipermail/speechd/attachments/20040415/ea3b5aa6/attachment.pgp From kirk@braille.uwo.ca Fri Apr 16 05:23:58 2004 From: Kirk Reiser To: speechd-discuss@nongnu.org Date: Sun Aug 13 12:52:46 2006 Subject: more questions as usual Message-ID: Hello everyone again: I have been reading through the various documents trying to gain a better understanding of how speech dispatcher works. There has been a couple things I haven't found answers for. The answers may be there but they've escaped me so far. Is there a way to change from one synthesizer module to another under the C API? I thought it might be useful to be able to change synths on the fly like it was suggested for various languages. Also there is some mention of work being done on a volume setting function so I wondered what the status on that might be? There is another function mentioned in the indices spd_commandline() which I cannot find any mention of in the main portion of the documentation though. I am willing to try my hand at writing the synth switching and volume functions at some point if they don't already exist. Are there any plans of including the speechd-up module in the debian speech-dispatcher file or would people prefer to have it made as a separate debian package? And, last but certainly not least, is there any interest for me to do some English fixes on the speech dispatcher documentation when I get the time and am reading through them. There are quite a few spelling mistakes and the sentence structure could be better in places. In general though it is very good. Kirk -- Kirk Reiser The Computer Braille Facility e-mail: kirk@braille.uwo.ca University of Western Ontario phone: (519) 661-3061 From pdm@freebsoft.org Fri Apr 16 08:31:34 2004 From: Milan Zamazal To: speechd-discuss@nongnu.org Date: Sun Aug 13 12:52:46 2006 Subject: more questions as usual In-Reply-To: (Kirk Reiser's message of "Thu, 15 Apr 2004 23:23:58 -0400") References: Message-ID: <87r7uocu2h.fsf@zamazal.org> >>>>> "KR" == Kirk Reiser writes: KR> Are there any plans of including the speechd-up module in the KR> debian speech-dispatcher file or would people prefer to have it KR> made as a separate debian package? Speechd-Up should be a separate package. Would you like to package it yourself? Regards, Milan Zamazal -- real programmer? don't get me started. if you need to hide your pathetic excuse for a carreer behind super-macho languages like C, C++, and/or Perl instead of writing clean, maintainable, efficient code, you aren't much of a real programmer in my view. -- Erik Naggum in comp.emacs From pdm@freebsoft.org Fri Apr 16 08:42:33 2004 From: Milan Zamazal To: speechd-discuss@nongnu.org Date: Sun Aug 13 12:52:46 2006 Subject: festival voices In-Reply-To: <1082045604.6787.8.camel@localhost> (Jon Hood's message of "Thu, 15 Apr 2004 16:13:25 +0000") References: <1081780885.6881.7.camel@localhost> <878yh0jtup.fsf@zamazal.org> <1082045604.6787.8.camel@localhost> Message-ID: <87llkwctk6.fsf@zamazal.org> >>>>> "JH" == Jon Hood writes: JH> I added the following to /usr/lib/festival/init.scm and set JH> "female1" in speechd.conf, but still no beans: You can try to telnet to the Festival server and check manually that everything works as expected there: telnet localhost 1314 (speechd-set-language "en") (speechd-set-voice "female1") (utt.play (speechd-speak* "Hello world.")) If the commands respond with LP (and some other information) and you can hear "hello world" in the expected voice, the Festival configuration is OK. Otherwise, if any of the commands responds with ER, investigate the Festival server log for more information about the error. BTW, don't forget you have to restart both Festival and Speech Dispatcher so that any Festival configuration change took its effect. The proper restarting sequence is: Stop Speech Dispatcher, restart Festival server, start Speech Dispatcher. Regards, Milan Zamazal -- SomeProgrammersLikeWritingLikeThis.However,IDontThinkThisFormOfCommunicationIs\ AGoodIdea.IApologizeToAllWhoCantReadMyTextsWrittenInATraditionalStyle. From kirk@braille.uwo.ca Fri Apr 16 12:52:00 2004 From: Kirk Reiser To: speechd-discuss@nongnu.org Date: Sun Aug 13 12:52:46 2006 Subject: more questions as usual In-Reply-To: <87r7uocu2h.fsf@zamazal.org> (Milan Zamazal's message of "Fri, 16 Apr 2004 08:31:34 +0200") References: <87r7uocu2h.fsf@zamazal.org> Message-ID: Well, I've been putting off learning how to build debian packages for quite a while now. I probably need to learn to do it myself! 'grin' You want to sponser me? 'wink' Kirk The lazier I get the more work I take on! -- Kirk Reiser The Computer Braille Facility e-mail: kirk@braille.uwo.ca University of Western Ontario phone: (519) 661-3061 From jon@squinky.gotdns.com Fri Apr 16 14:33:34 2004 From: Jon Hood To: speechd-discuss@nongnu.org Date: Sun Aug 13 12:52:46 2006 Subject: festival voices In-Reply-To: <87llkwctk6.fsf@zamazal.org> References: <1081780885.6881.7.camel@localhost> <878yh0jtup.fsf@zamazal.org> <1082045604.6787.8.camel@localhost> <87llkwctk6.fsf@zamazal.org> Message-ID: <1082118814.6775.7.camel@localhost> On Fri, 2004-04-16 at 06:42, Milan Zamazal wrote: > telnet localhost 1314 > (speechd-set-language "en") > (speechd-set-voice "female1") > (utt.play (speechd-speak* "Hello world.")) Thanks for the suggestions, I'm starting to make some progress. I have installed festival-freebsoft-utils, added "(require 'speech-dispatcher)" to /usr/lib/festival/init.scm, and now I don't get an error on (speechd-set-language "en"). However, # telnet localhost 1314 Trying 127.0.0.1... Connected to localhost. Escape character is '^]'. (speechd-set-language "en") LP ISO-8859-1 ft_StUfF_keyOK (speechd-set-voice "female1") ER (speechd-set-voice "male1") LP ISO-8859-1 ft_StUfF_keyOK The festival server says: SIOD ERROR: Undefined voice I have the following in speech-dispatcher.scm: (defvar speechd-language-voices '((english (male1 voice_kal_diphone) (male2 voice_ked_diphone) (female1 voice_us1_mbrola)) (americanenglish (male1 voice_kal_diphone) (male2 voice_ked_diphone)) (britishenglish (male1 voice_kal_diphone) (male2 voice_ked_diphone)) (german (male1 voice_german)) (czech (male1 voice_czech_mbrola_cz2))) # ls /usr/lib/festival/voices/english/ don_diphone ked_diphone us1_mbrola us3_mbrola kal_diphone rab_diphone us2_mbrola -Jon -- AIM: Squinky01 Yahoo!: Squinky86 ICQ: 160940989 Jabber: squinky86@im.gentoo.org -------------- next part -------------- A non-text attachment was scrubbed... Name: not available Type: application/pgp-signature Size: 189 bytes Desc: This is a digitally signed message part Url : http://lists.freebsoft.org/pipermail/speechd/attachments/20040416/2f727e5c/attachment.pgp From pdm@freebsoft.org Fri Apr 16 20:41:03 2004 From: Milan Zamazal To: speechd-discuss@nongnu.org Date: Sun Aug 13 12:52:46 2006 Subject: festival voices In-Reply-To: <1082118814.6775.7.camel@localhost> (Jon Hood's message of "Fri, 16 Apr 2004 12:33:34 +0000") References: <1081780885.6881.7.camel@localhost> <878yh0jtup.fsf@zamazal.org> <1082045604.6787.8.camel@localhost> <87llkwctk6.fsf@zamazal.org> <1082118814.6775.7.camel@localhost> Message-ID: <87n05bu5og.fsf@swan.zamazal.org> >>>>> "JH" == Jon Hood writes: JH> I have the following in speech-dispatcher.scm: JH> (defvar speechd-language-voices JH> '((english JH> (male1 voice_kal_diphone) JH> (male2 voice_ked_diphone) JH> (female1 voice_us1_mbrola)) JH> (americanenglish JH> (male1 voice_kal_diphone) JH> (male2 voice_ked_diphone)) JH> (britishenglish JH> (male1 voice_kal_diphone) JH> (male2 voice_ked_diphone)) JH> (german JH> (male1 voice_german)) JH> (czech JH> (male1 voice_czech_mbrola_cz2))) Define the `female1' voice for `americanenglish' and `britishenglish' as well. Currently, the Speech Dispatcher Festival interface takes the language from the Festival's `Language' parameter instead of the value set by Speech Dispatcher. This is a bug, I'll fix it. Setting the voice for the other English items as suggested above should help in the meantime. If it doesn't, look at the value of speechd-language-voices in the Festival session, whether it really contains the value set in init.scm. Regards, Milan Zamazal -- http://www.zamazal.org From pdm@freebsoft.org Fri Apr 16 20:59:21 2004 From: Milan Zamazal To: speechd-discuss@nongnu.org Date: Sun Aug 13 12:52:46 2006 Subject: more questions as usual In-Reply-To: (Kirk Reiser's message of "Fri, 16 Apr 2004 06:52:00 -0400") References: <87r7uocu2h.fsf@zamazal.org> Message-ID: <87hdvju4ty.fsf@swan.zamazal.org> >>>>> "KR" == Kirk Reiser writes: KR> You want to sponser me? 'wink' Sure. I could package Speechd-Up myself, but since I don't use Speakup, it could be hit by a significant level of incompetence. I think Debian packaging incompetence is easier to fix ;-), so I can sponsor your package if you promise it won't last forever and you get the courage to undergo the NMU process sometimes. :-| Regards, Milan Zamazal -- When you're in a fight with an idiot, it's difficult for other people to tell which one the idiot is. -- Bruce Perens in debian-devel From hanke@volny.cz Sat Apr 17 13:34:18 2004 From: Hynek Hanke To: speechd-discuss@nongnu.org Date: Sun Aug 13 12:52:46 2006 Subject: bug in installing libspeechd.h In-Reply-To: <20040416180154.17857.qmail@cesnet.brailcom.cz> References: <20040416180154.17857.qmail@cesnet.brailcom.cz> Message-ID: <20040417113418.GA1074@volny.cz> In the cvs changes message Kirk wrote: > Modified configure.in to add a test for > libspeechd.h which does not get installed by make install of speech > dispatcher. Thank you Kirk. > I don't know whether that should be considered a bug or > not. I suppose if speech dispatcher expects other applications to > link to it then it is a bug. Yes, it is a bug. I modified the corresponging Makefile.am to fix it. Regards, Hynek Hanke From kirk@braille.uwo.ca Sat Apr 17 16:11:17 2004 From: Kirk Reiser To: speechd-discuss@nongnu.org Date: Sun Aug 13 12:52:46 2006 Subject: bug in installing libspeechd.h In-Reply-To: <20040417113418.GA1074@volny.cz> (Hynek Hanke's message of "Sat, 17 Apr 2004 13:34:18 +0200") References: <20040416180154.17857.qmail@cesnet.brailcom.cz> <20040417113418.GA1074@volny.cz> Message-ID: Hi Hynek: Thanks for responding. I wasn't sure whether anyone read those cvs log messages other than myself. I still need to do a bit more with that automake/autoconf system because someone told me yesterday after I put my changes in place that configure said it couldn't find the libspeechd library but still compiled and worked. The amusing part is the libspeechd library was installed on their system but it may have been in a nonstandard place. So I definitely have more to learn about the autoconf system. Also, no one answered my questions about the missing functionality of speech dispatcher, switching synths on the fly and the other functions. I would just like to know whther I missed reading the documentation or currently you cannot switch synths. Kirk -- Kirk Reiser The Computer Braille Facility e-mail: kirk@braille.uwo.ca University of Western Ontario phone: (519) 661-3061 From hanke@volny.cz Sat Apr 17 16:22:54 2004 From: Hynek Hanke To: speechd-discuss@nongnu.org Date: Sun Aug 13 12:52:46 2006 Subject: bug in installing libspeechd.h In-Reply-To: References: <20040416180154.17857.qmail@cesnet.brailcom.cz> <20040417113418.GA1074@volny.cz> Message-ID: <20040417142254.GA20357@volny.cz> On Sat, Apr 17, 2004 at 10:11:17AM -0400, Kirk Reiser wrote: > Hi Hynek: Thanks for responding. I wasn't sure whether anyone read > those cvs log messages other than myself. Yes, I do. You should put the same message into the Changelog also. Hynek From hanke@volny.cz Sat Apr 17 17:33:49 2004 From: Hynek Hanke To: speechd-discuss@nongnu.org Date: Sun Aug 13 12:52:46 2006 Subject: more questions as usual In-Reply-To: References: Message-ID: <20040417153349.GB20357@volny.cz> On Thu, Apr 15, 2004 at 11:23:58PM -0400, Kirk Reiser wrote: > Is there a way to change from one synthesizer module to another under > the C API? I thought it might be useful to be able to change synths > on the fly like it was suggested for various languages. Also there is > some mention of work being done on a volume setting function so I > wondered what the status on that might be? These functions are available in SSIP (SET VOLUME since version 0.3) but were not available in the C API. I've included these two in the CVS version now. I've also made public the functions that I use to send arbitrary strings through the SSIP channell so that the programmer is able to use a SSIP command through the library even though it's still not included in the C API. > There is another function > mentioned in the indices spd_commandline() which I cannot find any > mention of in the main portion of the documentation though. From jschmude@adelphia.net Sun Apr 25 22:27:59 2004 From: Jacob Schmude To: speechd-discuss@nongnu.org Date: Sun Aug 13 12:52:46 2006 Subject: odd behavior with speech dispatcher CVS Message-ID: Hi all I just updated my speech dispatcher from CVS at around 4:00 PM EDT today. I'm noticing an odd bug which occurs with all synth modules. What is happening is that lines are running together. I'm using speech dispatcher with speechd-up, and when I issue a command that reads more than one line, the last word of the currently speaking line gets mashed together with the first word of the line after it. This causes some pronunciation anomolies. This didn't happen with the slightly older CVS build I was using before. I've been able to reproduce this behavior using flite, festival, and dectalk software. Has anyone else noticed this slightly odd behavior? From hanke@volny.cz Mon Apr 26 00:16:14 2004 From: Hynek Hanke To: speechd-discuss@nongnu.org Date: Sun Aug 13 12:52:46 2006 Subject: odd behavior with speech dispatcher CVS In-Reply-To: References: Message-ID: <20040425221614.GA4977@volny.cz> > I just updated my speech dispatcher from CVS at around 4:00 PM EDT today. > I'm noticing an odd bug which occurs with all synth modules. What is > happening is that lines are running together. I'm using speech dispatcher > with speechd-up, and when I issue a command that reads more than one line, > the last word of the currently speaking line gets mashed together > with the first word of the line after it. This causes some pronunciation > anomolies. This didn't happen with the slightly older CVS build I was > using before. I've been able to reproduce this behavior using flite, > festival, and dectalk software. Hi Jacob, either I can't reproduce the problem or I don't recognize it. Please could you enable Festival Speech Dispatcher output module debugging in etc/speech-dispatcher/modules/festival.conf with the options something like Debug 1 and DebugFile "/tmp/festival-debug.log" reproduce the error and send me the festival-debug.log file together with the string it failed to read correctly? This way, I'll be able too look at what string exactly was received from Speechd-Up and what string was passed on to Festival. Thank you, Hynek From chuckh@sent.com Wed Apr 28 18:19:13 2004 From: Chuck Hallenbeck To: speechd-discuss@nongnu.org Date: Sun Aug 13 12:52:46 2006 Subject: help with speechd-up please Message-ID: -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 I have a configuration and compilation failure when installing speechd-up on this Slackware 9.1 system. I have previously successfully installed flite and speech-dispatcher, and the "basic.test" worked as expected, with pitch and rate changes. Festival is not installed. The failure is the inability to locate "libspeechd.h" as first reported when configuring speechd-up, and then of course it prevents the "make all" from working. That file actually is present though, in this location: /usr/local/src/speech-dispatcher-0.3/src/c/api/libspeechd.h I have tried to follow the installation steps with care, but I must have overlooked something. Can you please advise? Chuck - -- The Moon is Waxing Gibbous (59% of Full) My home page is now at http://www.mhcable.com/~chuckh -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.2.4 (GNU/Linux) iQCVAwUBQI/ZmDVdG8M9x9tGAQLzwAP/aQ0TvVIxbbzs937DNddNEdIoVIXI3c6/ 3qQ/I7aKfOjUBWferfpKD+X/MBlDIk1ijDR4e5uJy+M7Phl1JiFHeQQSs5aEWOHF 21h2H+q4P5dwdmTjO2dyTye9kWIVGNnOWgXbzjC2/ypdWDuOZClQ6msa0T30pZaR Zm5pDkKMMOE= =k7X7 -----END PGP SIGNATURE----- From pdm@freebsoft.org Tue May 11 14:53:48 2004 From: Milan Zamazal To: speechd-discuss@nongnu.org Date: Sun Aug 13 12:52:46 2006 Subject: Announce: festival-freebsoft-utils 0.2 released Message-ID: <87vfj3untv.fsf@zamazal.org> festival-freebsoft-utils 0.2 released ===================================== The Free(b)soft project is pleased to announce festival-freebsoft-utils 0.2. * What is festival-freebsoft-utils? festival-freebsoft-utils is a collection of Festival utilities that enhance Festival with some useful features. They provide all what is needed for interaction with Speech Dispatcher. Key festival-freebsoft-utils features are: - Generalized concept of input events. festival-freebsoft-utils allows not only plain text synthesis, but also combining it with sounds. Additionally, mechanism of logical events mapped to other events is provided. - Substitution of events for given words. - High-level voice selection mechanism and setting of basic prosodic parameters. - Spelling mode. - Capital letter signalization. - Punctuation modes, for explicit reading or not reading punctuation characters. - Incremental synthesis of texts and events. - Speech Dispatcher support. - Rudimentary SSML support. - Enhancing the Festival extension language with some functions commonly used in Lisp. Support for wrapping already defined Festival functions by your own code. - Everything is written in the extension language, no patching of the Festival C++ sources is needed. * What is new in the 0.2 version? - Event handling mechanism was completely reworked to be cleaner and more robust. - Voice and language selection mechanism was generalized and simplified. - Incremental synthesis of texts and events, including utterance chunking implemented in the extension language. - Rudimentary SSML support. - New functions for unified setting of prosodic parameters and avoiding initial spaces. - New utility functions. * Where to get it? You can get the distribution tarball of the released version from http://www.freebsoft.org/pub/projects/festival-freebsoft-utils/festival-freebsoft-utils-0.2.tar.gz . The home page of the project is http://www.freebsoft.org/festival-freebsoft-utils . From tcross@rapttech.com.au Sat May 15 15:05:40 2004 From: Tim Cross To: speechd-discuss@nongnu.org Date: Sun Aug 13 12:52:46 2006 Subject: speechd modules Message-ID: <16550.5540.249263.109171@tiger.rapttech.com.au> Hi All, I'm new to speechd and have to say I like the concept and implementation as far as I've seen it so far. I'm thinking that in my spare time, I wouldn't mind adding some new output modules for speechd. In particular, I thought I might look at doing some modules for the software dectalk and possibly Cepstral's TTS engine. However, I thought before I begin to look at this, I'd see if anyone else was working on anything with these software synths as I'd prefer to work with someone rather than reinvent what they have already done. I realise speechd is pro open source and neither of these synths are open source. However, I also feel we need some modules with high quality speech and there is some validity in supporting commercial efforts with TTS, especially if the retail price is reasonable. Just for some background on me. I've been working as a programmer for about 16 years and a Linux user for the past 10. I lost my vision about 7 ytears ago and have been using emacspeak since then. I'm very interested in the speechd-el as an additional alternative approach. While I totally support and commend T.V. Raman for the work he has done, I feel emacspeak has grown a little bit large and difficult to manage. In fact, I've been tossing up the idea of re-writing parts of it to simplify parts etc. However, speechd-el could offer an alternative approach where we could try implementing some of the nice features of emacspeak in a relatively simple way without major re-writes etc. I'm not trying to be critical of emacspeak in any way - the issues I have with it are just a reflection of the type of issues you will find in any system which has grown and evolved over a long period - especially one that has, to a large extent, pushed the boundaries of traditional speech systems. I'm not interested in making speechd-el a clone of emacpseak, but there are some features in emacpseak which I think wold be valuable additions to speechd-el. One area where emacspeak is very much lacking is in its interface to speech synthesizes. There are few support synths and its a bit tricky to add new ones. This is in particular what I like about the speechd approach. I think the abstraction being used is a great idea and offers a lot of possiblities. Having said all of this, I should point out my time is quite limited and it will take me a while to actually make much of a contribution to either speechd or speeched-el. I made a slight tactical error in my employment about 8 months ago and took on a 2 year position managing a reasonably large data centre. This is now taking most of my time. However, I miss coding and like to still spend some free time hacking away at code and this is why I'd like to work on some project - albeit very slowly. so, I'd like to here if my ideas and desire to contribute will fit into the philosophy of speechd and if anyone else is working on either software dectalk or Cepstral support for speechd. Tim From jschmude@adelphia.net Sat May 15 16:49:23 2004 From: Jacob Schmude To: speechd-discuss@nongnu.org Date: Sun Aug 13 12:52:46 2006 Subject: speechd modules In-Reply-To: <16550.5540.249263.109171@tiger.rapttech.com.au> References: <16550.5540.249263.109171@tiger.rapttech.com.au> Message-ID: Hi I've got software dectalk working with speech dispatcher. The approach I've taken is to utilize the generic module to call DECTalk's say command, with appropriate parameters for pitch, rate, and voice. Generally, I find this to work quite well, but a native C module would no doubt be better. However, I just don't have the time to do it at the moment, and programming for dectalk isn't the easiest thing in the world, especially for a beginner programmer like myself. I took the generic approach because I really needed good software speech rather quickly. I believe there's someone on the speakup list working on a Cepstral module as well, though I can't remember who it is at the moment. On Sat, 15 May 2004, Tim Cross wrote: > Hi All, > > I'm thinking that in my spare time, I wouldn't mind adding some new > output modules for speechd. In particular, I thought I might look at > doing some modules for the software dectalk and possibly Cepstral's > TTS engine. However, I thought before I begin to look at this, I'd see > if anyone else was working on anything with these software synths as > I'd prefer to work with someone rather than reinvent what they have > already done. From pdm@freebsoft.org Sun May 16 14:59:39 2004 From: Milan Zamazal To: speechd-discuss@nongnu.org Date: Sun Aug 13 12:52:46 2006 Subject: speechd-el features (was: speechd modules) In-Reply-To: <16550.5540.249263.109171@tiger.rapttech.com.au> (Tim Cross's message of "Sat, 15 May 2004 23:05:40 +1000") References: <16550.5540.249263.109171@tiger.rapttech.com.au> Message-ID: <87oeoozfwk.fsf@zamazal.org> >>>>> "TC" == Tim Cross writes: TC> I'm not interested in making speechd-el a clone of emacpseak, Sure, speechd-el and Emacspeak solve the same problem, but in a very different way; it would make no sense to mutually clone them. TC> but there are some features in emacpseak which I think wold be TC> valuable additions to speechd-el. Maybe it would be useful to mention the particular features? If they fit into speechd-el, I can create a to-do list for people who would like to contribute to speechd-el. Regards, Milan Zamazal -- This mail contains a hashcash stamp. See http://www.hashcash.org for more information. From tcross@rapttech.com.au Mon May 17 00:22:51 2004 From: Tim Cross To: speechd-discuss@nongnu.org Date: Sun Aug 13 12:52:46 2006 Subject: speechd-el features (was: speechd modules) In-Reply-To: <87oeoozfwk.fsf@zamazal.org> References: <16550.5540.249263.109171@tiger.rapttech.com.au> <87oeoozfwk.fsf@zamazal.org> Message-ID: <16551.59835.766591.790732@tiger.rapttech.com.au> Sounds good. I'll use it for a while - possibly while I try to develop some other output modules and once I've got use to speechd-el to a greater extent, I'll make some suggestions. Actually, so far, I'm pretty impressed with speechd-el. It seems a bit less intrusive and I've found a few areas were it works better in that it doesn't have some minor "bugs" I've run into with emacspeak. What I really like is the whole speechd concept and the abstraction from the synthesizer level. I think this offers a lot of potential. One area I was thinking about this morning was integrating speechd support into the sawfish window manager. While I personally would be quite happy without any window manager, places I work often have no choice but to use a X session and some speech support for the window manager would be very nice. Tim >>>>> "Milan" == Milan Zamazal writes: >>>>> "TC" == Tim Cross writes: TC> I'm not interested in making speechd-el a clone of emacpseak, Milan> Sure, speechd-el and Emacspeak solve the same problem, but in Milan> a very different way; it would make no sense to mutually clone Milan> them. TC> but there are some features in emacpseak which I think wold be TC> valuable additions to speechd-el. Milan> Maybe it would be useful to mention the particular features? Milan> If they fit into speechd-el, I can create a to-do list for Milan> people who would like to contribute to speechd-el. Milan> Regards, Milan> Milan Zamazal Milan> -- This mail contains a hashcash stamp. See Milan> http://www.hashcash.org for more information. From pdm@freebsoft.org Mon May 17 20:32:07 2004 From: Milan Zamazal To: speechd-discuss@nongnu.org Date: Sun Aug 13 12:52:46 2006 Subject: Speaking window manager (was: speechd-el features) In-Reply-To: <16551.59835.766591.790732@tiger.rapttech.com.au> (Tim Cross's message of "Mon, 17 May 2004 08:22:51 +1000") References: <16550.5540.249263.109171@tiger.rapttech.com.au> <87oeoozfwk.fsf@zamazal.org> <16551.59835.766591.790732@tiger.rapttech.com.au> Message-ID: <87k6zaykew.fsf_-_@zamazal.org> >>>>> "TC" == Tim Cross writes: TC> One area I was thinking about this morning was integrating TC> speechd support into the sawfish window manager. We thought about this once too. From the technical point view, it shouldn't be difficult to write Speech Dispatcher support for Sawfish, it's possible to port the Elisp or Common Lisp interface. But the question is, whether this is the right approach. For instance, Gnopernicus has some window manager support -- wouldn't be better to use that? Or can Sawfish extensibility be used in some advanced way to better serve the needs of visually impaired in the windowing environment? I don't know answers to these questions. Regards, Milan Zamazal -- Why waste keystrokes when you can waste CPU cycles instead? Karl Fogel From pdm@freebsoft.org Thu May 27 12:02:48 2004 From: Milan Zamazal To: speechd-discuss@nongnu.org Date: Sun Aug 13 12:52:46 2006 Subject: speechd-el 0.4 released Message-ID: <87ekp62nnb.fsf@zamazal.org> speechd-el 0.4 released ======================= The Free(b)soft project is pleased to announce speechd-el 0.4. * What is speechd-el? speechd-el is an Emacs client to Speech Dispatcher, providing a complex speech interface to Emacs, focused especially on (but not limited to) the blind and visually impaired users. It allows the user to work with Emacs without looking on the screen, using the speech output produced by the synthesizers supported in Speech Dispatcher. Key speechd-el features are: - Automated speaking of most Emacs messages and actions, without necessity to customize each particular Emacs function to make it speak. - Most built-in and external Emacs packages can produce speech output immediately, without any special support. - Almost no change of the standard Emacs behavior, no interference with user settings. - Highly configurable. - Full cooperation with Speech Dispatcher. - Small code size. - Free Software, distributed under the GPL. The speechd-el package contains an Elisp library for communication with Speech Dispatcher, allowing you to write your own Emacs speech output extensions and your own Emacs interfaces to Speech Dispatcher. * What is new in the 0.4 version? - Whitespace signalization. - Repetition of last speech output possible in character reading prompts. - Language selection based on currently selected input method. - Asynchronous action handling reworked. - Responsiveness improved. See the NEWS file for more details. * Where to get it? You can get the distribution tarball of the released version from http://www.freebsoft.org/pub/projects/speechd-el/speechd-el-0.4.tar.gz . The home page of the project is http://www.freebsoft.org/speechd-el . From hanke@volny.cz Fri May 28 13:32:31 2004 From: Hynek Hanke To: speechd-discuss@nongnu.org Date: Sun Aug 13 12:52:46 2006 Subject: speechd Message-ID: <20040528113231.GD1345@volny.cz> Speech Dispatcher 0.4 released ============================== The Brailcom organization is happy to announce the release of Speech Dispatcher 0.4 developed as a part of the Free(b)Soft project. * What is Speech Dispatcher? Speech Dispatcher is a device independent layer for speech synthesis, developed with the goal of making the usage of speech synthesis easier for application programmers. It takes care of most of the tasks necessary to solve in speech enabled applications. What is a very high level GUI library to graphics, Speech Dispatcher is to speech synthesis. The architecture of Speech Dispatcher is based on a proven client/server model. The basic means of client communication with Speech Dispatcher is through a TCP connection using the Speech Synthesis Interface Protocol (SSIP). Key Speech Dispatcher features are: - Message priority model that allows multiple simultaneous connections to Speech Dispatcher from one or more clients and tries to provide the user with the most important messages. - Different output modules that talk to different synthesizers so that the programmer doesn't need to care which particular synthesizer is being used. Currently Festival, Flite, Epos and (non-free) Dectalk software are supported. - Client-based configuration allows users to configure different settings for different clients that connect to Speech Dispatcher. - Simple interface for programs written in C, C++ provided through a shared library and a Common Lisp interface. The Python interface is being worked on. An Elisp library is developed as a sperate project speechd-el. Possibly an interface to any other language can be developed. * What is new in 0.4? - Flite output is now installed with the 16-bit voice, if available. - The generic output module can now also set volume. - spd-say is now rewritten to accepts the most important speech parameters and thus provide a simple but useful Speech Dispatcher client. Documentation is provided in info. - The C/C++ shared library interface now uses TCP_NODELAY to prevent timeouts and make the communication faster. - Bugfixes. * Where to get it? You can get the distribution tarball of the released version from http://www.freebsoft.org/pub/projects/speechd/speech-dispatcher-0.4.tar.gz . We recommend you to fetch the sound icons Speech Dispatcher can use as well. They are available at http://www.freebsoft.org/pub/projects/sound-icons/sound-icons-0.1.tar.gz Corresponding Debian packages are or soon will be available at your Debian distribution mirror. (We would highly appreciate any help with RPM distribution!) The home page of the project is http://www.freebsoft.org/speechd . * How to report bugs? Please report bugs at . For other contact please use Happy synthesizing! From jschmude@adelphia.net Sat May 29 15:57:20 2004 From: Jacob Schmude To: speechd-discuss@nongnu.org Date: Sun Aug 13 12:52:46 2006 Subject: problem with generic module? Message-ID: Hi all I just upgraded to speech dispatcher 0.4. Previously, I had no problems using the generic module to output to software dectalk. After the upgrade, and resetting the parameters in the config file to their appropriate values, I get no speech. The debug log for the module reveals: Sat May 29 09:24:54 2004 [906025]: Generic: creating new thread for generic_speak Sat May 29 09:24:54 2004 [906279]: generic: speaking thread starting....... Sat May 29 09:24:56 2004 [457878]: write() It almost looks like none of the text is getting through to the module, as those are the last lines of the log file. Further, the speechd log file yields: [Sat May 29 09:42:28 2004] speechd: Module flite loaded. [Sat May 29 09:42:28 2004] speechd: Module festival loaded. [Sat May 29 09:42:28 2004] speechd: Module epos-generic loaded. [Sat May 29 09:42:28 2004] speechd: Module dtk-generic loaded. [Sat May 29 09:42:28 2004] speechd: Configuration has been read from "/usr/local/etc/speech-dispatcher//speechd.conf" [Sat May 29 09:42:28 2004] speechd: Speech Dispatcher started with 4 output modules [Sat May 29 09:42:28 2004] speechd: Openning a socket connection [Sat May 29 09:42:28 2004] speechd: Speech Dispatcher waiting for clients ... [Sat May 29 09:42:30 2004] speechd: Adding client on fd 17 [Sat May 29 09:42:30 2004] speechd: Output locked... [Sat May 29 09:42:30 2004] speechd: Removing client on fd 17 [Sat May 29 09:42:30 2004] speechd: Connection closed [Sat May 29 09:42:30 2004] speechd: Broken pipe to module. [Sat May 29 09:42:30 2004] speechd: Output module working status: 0 (pid:4281)[Sat May 29 09:42:30 2004] speechd: Output module terminated abnormally, probably crashed. [Sat May 29 09:42:30 2004] speechd: Output unlocked. [Sat May 29 09:42:30 2004] speechd: Error: Output module failed [Sat May 29 09:42:30 2004] speechd: Output module working status: 0 (pid:0) [Sat May 29 09:42:30 2004] speechd: Output module terminated abnormally, probably crashed. I've looked over the necessary config files and I'm sure everything is set properly. Has anyone else seen this? Is there a new format for generic config files? This occurs both when trying to use spd-say and speechd_up. Flite works fine, however. Any ideas on this one? From hanke@volny.cz Sat May 29 16:13:44 2004 From: Hynek Hanke To: speechd-discuss@nongnu.org Date: Sun Aug 13 12:52:47 2006 Subject: speechd modules In-Reply-To: <16550.5540.249263.109171@tiger.rapttech.com.au> References: <16550.5540.249263.109171@tiger.rapttech.com.au> Message-ID: <20040529141344.GA1620@volny.cz> On Sat, May 15, 2004 at 11:05:40PM +1000, Tim Cross wrote: > I thought I might look at > doing some modules for the software dectalk and possibly Cepstral's > TTS engine. However, I thought before I begin to look at this, I'd see > if anyone else was working on anything with these software synths as > I'd prefer to work with someone rather than reinvent what they have > already done. Hi Tim, thank you for your message and please forgive my late response. There currently *is* some very simple output module for software Dectalk. It works using the generic output module (the configuration file is named dtk-generic.conf and can be found in etc/speech-dispatcher/modules/. So, if you have Dectalk installed, loading the synthesizer dtk-generic should make it speak. However, the generic output module is a very simple interface (it lacks support for punctuation modes, etc.). So, if you like to help with this, you might investigate if there isn't some better way of communication than it's simple line client. The current config file dtk-generic.conf was written by Jacob Schmude , a Speakup developer. As for Cepstral, somebody on the Speakup mailing list wanted to create a new config file for the generic output module, but I didn't hear anything from him for a longer time now so I assume nobody is working on it. Your work on the Cepstral module might have more impact on the end users than improving the Dectalk software synthesis module. > I realise speechd is pro open source and neither of these synths are > open source. However, I also feel we need some modules with high > quality speech and there is some validity in supporting commercial > efforts with TTS, especially if the retail price is reasonable. I think we are not only pro Open Source, but also pro Free Software ;) If you can help with Festival (which is Free Software), this would be highly appreciated. On the other side, we understand that e.g. the Cepstral voices might make some users whose language is not supported yet in Festival able to use the GNU/Linux system at all. So in the area of software synthesis, we decided we have to temporarily suppport even non-free software although we don't like it very much. I'm happy to help you with anything that was proposed here. With Regards, Hynek From hanke@volny.cz Sat May 29 18:51:28 2004 From: Hynek Hanke To: speechd-discuss@nongnu.org Date: Sun Aug 13 12:52:47 2006 Subject: problem with generic module? In-Reply-To: References: Message-ID: <20040529165128.GA3671@volny.cz> On Sat, May 29, 2004 at 09:57:20AM -0400, Jacob Schmude wrote: > I just upgraded to speech dispatcher 0.4. Previously, I had no > problems using the generic module to output to software dectalk. After the > upgrade, and resetting the parameters in the config file to their > appropriate values, I get no speech. Arrrgh... The module is broken, my fault. Because I don't use dectalk, I thought I would send it to you before the release so that you add the volume option to dtk-generic.conf and, in the process, you will test if everything works correctly after the updates I've made to the module. But in the meantime, we decided not to include the volume support, so you didn't test the module. It didn't occur to me that you actually had no reason to run it, so I assumed everything is ok and didn't test it either. Can you please try the current CVS version if everything works correctly? It includes the latest dtk-generic.conf file you have sent me. Now, I'm not sure what to do. Should we make a 0.4.1 bugfix release? 0.4 should work for most users, but for those that use DecTalk, it's broken :( With Regards, Hynek Hanke From hanke@volny.cz Sun May 30 12:07:22 2004 From: Hynek Hanke To: speechd-discuss@nongnu.org Date: Sun Aug 13 12:52:47 2006 Subject: Speech Dispatcher 0.4.1 released Message-ID: <20040530100722.GE1280@volny.cz> Speech Dispatcher 0.4.1 released ================================ The Brailcom organization is happy to announce the release of Speech Dispatcher 0.4.1 developed as a part of the Free(b)Soft project. * What is Speech Dispatcher? Speech Dispatcher is a device independent layer for speech synthesis, developed with the goal of making the usage of speech synthesis easier for application programmers. It takes care of most of the tasks necessary to solve in speech enabled applications. What is a very high level GUI library to graphics, Speech Dispatcher is to speech synthesis. The architecture of Speech Dispatcher is based on a proven client/server model. The basic means of client communication with Speech Dispatcher is through a TCP connection using the Speech Synthesis Interface Protocol (SSIP). Key Speech Dispatcher features are: - Message priority model that allows multiple simultaneous connections to Speech Dispatcher from one or more clients and tries to provide the user with the most important messages. - Different output modules that talk to different synthesizers so that the programmer doesn't need to care which particular synthesizer is being used. Currently Festival, Flite, Epos and (non-free) Dectalk software are supported. - Client-based configuration allows users to configure different settings for different clients that connect to Speech Dispatcher. - Simple interface for programs written in C, C++ provided through a shared library and a Common Lisp interface. The Python interface is being worked on. An Elisp library is developed as a sperate project speechd-el. Possibly an interface to any other language can be developed. * What is new in 0.4.1? This is a bugfix release because there was a severe bug in the 0.4 release in the Generic output module. Below follows the list of new features in 0.4: - Flite output is now installed with the 16-bit voice, if available. - The generic output module can now also set volume. - spd-say is now rewritten to accept the most important speech parameters and thus provide a simple but useful Speech Dispatcher client. Documentation is provided in info. - The C/C++ shared library interface now uses TCP_NODELAY to prevent timeouts and make the communication faster. - Bugfixes. * Where to get it? You can get the distribution tarball of the released version from http://www.freebsoft.org/pub/projects/speechd/speech-dispatcher-0.4.1.tar.gz . We recommend you to fetch the sound icons Speech Dispatcher can use as well. They are available at http://www.freebsoft.org/pub/projects/sound-icons/sound-icons-0.1.tar.gz Corresponding Debian packages are or soon will be available at your Debian distribution mirror. (We would highly appreciate any help with RPM distribution!) The home page of the project is http://www.freebsoft.org/speechd . * How to report bugs? Please report bugs at . For other contact please use Happy synthesizing! From nath.ml@free.fr Sun Jun 6 09:36:19 2004 From: Nath To: speechd-discuss@nongnu.org Date: Sun Aug 13 12:52:47 2006 Subject: French language with Speech Dispatcher, Speechd-el and DECtalk software 5.0 Message-ID: <878yf1rvd8.fsf@spy.home> Hello, First excuse me if this message appears two times on the mailing list. I explain : yesterday I posted this question directly on the web interface (without subscribtion) but it seems it didn't work because this morning I can't see my post on the archives (no messages for june are displayed). So my question is : i installed Speech dispatcher + Speechd-el + DECtalk software 5.0 : all seems to wwork fine but in english only. I have files for DECtalk french language but I don't find how to indicate to Speech Dispatcher to use these files in order to have french language enabled. So, all your ideas & suggestions are very welcome. Thx in advance for your help, -- Nath From hanke@volny.cz Tue Jun 8 19:17:20 2004 From: Hynek Hanke To: speechd-discuss@nongnu.org Date: Sun Aug 13 12:52:47 2006 Subject: French language with Speech Dispatcher, Speechd-el and DECtalk software 5.0 In-Reply-To: <878yf1rvd8.fsf@spy.home> References: <878yf1rvd8.fsf@spy.home> Message-ID: <20040608171720.GE1280@volny.cz> On Sun, Jun 06, 2004 at 09:36:19AM +0200, Nath wrote: > i installed Speech dispatcher + Speechd-el + DECtalk software 5.0 : all > seems to wwork fine but in english only. > I have files for DECtalk french language but I don't find how to > indicate to Speech Dispatcher to use these files in order to have french > language enabled. First of all, excuse my late response. I was busy with other things. It *is* possible to set the language parameter of the Generic output module that we use for Dectalk. However, the current dectalk configureation file for the module doesn't support it (it isn't implemented) because Jacob, who was writing the module, is an English native speaker. So, switching languages should be quite trivial, but I currently don't know how to do it because I don't have a Dectalk software synthesizer on my machine (it's proprietary after all). Could you plese tell me how does one set a different language with the "say" line client that is provided with DecTalk? I might then tell you what you have to change in the configuration file. Or does Jacob know something about it? With Regards, Hynek From nath.ml@free.fr Tue Jun 8 21:36:09 2004 From: Nath To: speechd-discuss@nongnu.org Date: Sun Aug 13 12:52:47 2006 Subject: French language with Speech Dispatcher, Speechd-el and DECtalk software 5.0 In-Reply-To: <20040608171720.GE1280@volny.cz> (Hynek Hanke's message of "Tue, 8 Jun 2004 19:17:20 +0200") References: <878yf1rvd8.fsf@spy.home> <20040608171720.GE1280@volny.cz> Message-ID: <87n03d266u.fsf@spy.home> Hynek Hanke writes: > On Sun, Jun 06, 2004 at 09:36:19AM +0200, Nath wrote: >> i installed Speech dispatcher + Speechd-el + DECtalk software 5.0 : all >> seems to wwork fine but in english only. >> I have files for DECtalk french language but I don't find how to >> indicate to Speech Dispatcher to use these files in order to have french >> language enabled. > > First of all, excuse my late response. I was busy with other things. > No problem ! > It *is* possible to set the language parameter of the Generic output > module that we use for Dectalk. However, the current dectalk > configureation file for the module doesn't support it (it isn't > implemented) because Jacob, who was writing the module, is an English > native speaker. > > So, switching languages should be quite trivial, but I currently don't > know how to do it because I don't have a Dectalk software synthesizer > on my machine (it's proprietary after all). Could you plese tell me > how does one set a different language with the "say" line client > that is provided with DecTalk? I might then tell you what you have > to change in the configuration file. > > Or does Jacob know something about it? > Ok. The problem is that I tried to find how to tell the "say" program to use french but unfortunatly don't find the solution for the moment. so, i'll search again and let you know as soon as I found someting. > With Regards, > Hynek > Thx for your response and best regards, -- Nath From philsfree@free.fr Wed Jun 9 00:46:26 2004 From: Phil's Free To: speechd-discuss@nongnu.org Date: Sun Aug 13 12:52:47 2006 Subject: French language with Speech Dispatcher, Speechd-el and DECtalk software 5.0 In-Reply-To: <87n03d266u.fsf@spy.home> References: <878yf1rvd8.fsf@spy.home> <20040608171720.GE1280@volny.cz> <87n03d266u.fsf@spy.home> Message-ID: <1086734786.4537.44.camel@toinette.vif.argent.fr> Le mar 08/06/2004 ? 21:36, Nath a ?crit : > Hynek Hanke writes: > > > On Sun, Jun 06, 2004 at 09:36:19AM +0200, Nath wrote: > >> i installed Speech dispatcher + Speechd-el + DECtalk software 5.0 : all > >> seems to wwork fine but in english only. > >> I have files for DECtalk french language but I don't find how to > >> indicate to Speech Dispatcher to use these files in order to have french > >> language enabled. > > > > First of all, excuse my late response. I was busy with other things. > > > No problem ! > > > It *is* possible to set the language parameter of the Generic output > > module that we use for Dectalk. However, the current dectalk > > configureation file for the module doesn't support it (it isn't > > implemented) because Jacob, who was writing the module, is an English > > native speaker. > > > > So, switching languages should be quite trivial, but I currently don't > > know how to do it because I don't have a Dectalk software synthesizer > > on my machine (it's proprietary after all). Could you plese tell me > > how does one set a different language with the "say" line client > > that is provided with DecTalk? I might then tell you what you have > > to change in the configuration file. > > > > Or does Jacob know something about it? > > > Ok. The problem is that I tried to find how to tell the "say" program to > use french but unfortunatly don't find the solution for the moment. so, > i'll search again and let you know as soon as I found someting. > To find out any convenient option have you tried : say --help say -h say -? strings $(which say) | grep -i lang strings $(which say) | less My 2 cents. -- Phil From gcasse@oralux.org Thu Jun 10 13:26:49 2004 From: Gilles Casse To: speechd-discuss@nongnu.org Date: Sun Aug 13 12:52:47 2006 Subject: French language with Speech Dispatcher, Speechd-el and DECtalk software 5.0 Message-ID: <16584.17785.22678.280137@gargle.gargle.HOWL> Hello, I fear we can't select the language using the original say software from Fonix. So I have rewritten it so that Nath can use it in french via Speech-Dispatcher. A new option has been added to select the language (e.g. for spanish "-l es" ). At the moment, the default language is french :-). However, I meet an issue when typing french characters under Emacs: For example, the c cedilla (0xe7) seems to crash the output module. When I type: fran?ais each character is spelled until the 'n' Files dtk-speak.txt [1] and /var/log/speechd.log [2] are copied at the end of this post. If the person who worked on the DECtalk software integration is interested in testing other languages, I am sure Fonix will send them to him for free. Since I have been told that their politic is to supply the additional expected languages, once the core file has been bought. Thanks for your help, Gilles [1] /tmp/dtk-speak.txt [:np][:ra 338][:dv ap 225] n [2] /var/log/speechd.log [Thu Jun 10 12:56:17 2004] speechd: Output locked... [Thu Jun 10 12:56:17 2004] speechd: Output unlocked. [Thu Jun 10 12:56:17 2004] speechd: Output locked... [Thu Jun 10 12:56:17 2004] speechd: Output locked... [Thu Jun 10 12:56:17 2004] speechd: Broken pipe to module. [Thu Jun 10 12:56:17 2004] speechd: Output module working status: 0 (pid:28112) [Thu Jun 10 12:56:17 2004] speechd: Output module terminated abnormally, probably crashed. [Thu Jun 10 12:56:17 2004] speechd: Output unlocked. [Thu Jun 10 12:56:17 2004] speechd: Error: Output module failed [Thu Jun 10 12:56:17 2004] speechd: Output module working status: 3 (pid:0) [Thu Jun 10 12:56:17 2004] speechd: Broken pipe to module. [Thu Jun 10 12:56:17 2004] speechd: Output module working status: 0 (pid:28112) [Thu Jun 10 12:56:17 2004] speechd: Output module terminated abnormally, probably crashed. [Thu Jun 10 12:56:17 2004] speechd: Output unlocked. [Thu Jun 10 12:56:17 2004] speechd: Output locked... [Thu Jun 10 12:56:17 2004] speechd: Output unlocked. [Thu Jun 10 12:56:17 2004] speechd: Output locked... [Thu Jun 10 12:56:17 2004] speechd: Can't find default output module or it's not working! [Thu Jun 10 12:56:17 2004] speechd: Broken pipe to module. [Thu Jun 10 12:56:17 2004] speechd: Output module working status: 0 (pid:28112) [Thu Jun 10 12:56:17 2004] speechd: Output module terminated abnormally, probably crashed. [Thu Jun 10 12:56:17 2004] speechd: Output unlocked. [Thu Jun 10 12:56:17 2004] speechd: Error: Output module failed [Thu Jun 10 12:56:17 2004] speechd: Output module working status: 3 (pid:0) -- Oralux http://oralux.org From pdm@freebsoft.org Wed Jun 16 20:00:57 2004 From: Milan Zamazal To: speechd-discuss@nongnu.org Date: Sun Aug 13 12:52:47 2006 Subject: Guile interface to SSIP Message-ID: <871xkf4c2u.fsf@zamazal.org> Simple Guile interface to SSIP is now available in the Speech Dispatcher CVS, in the src/guile/ subdirectory. See http://www.freebsoft.org/speechd for information how to access Speech Dispatcher CVS. (Hint: There is already available a simple AT-SPI Guile interface by Mario Lang, http://www.delysid.org/gspi.html.) Regards, Milan Zamazal -- "Having GNU Emacs is like having a dragon's cave of treasures." Robert J. Chassell From pdm@freebsoft.org Fri Jul 2 11:33:38 2004 From: Milan Zamazal To: speechd-discuss@nongnu.org Date: Sun Aug 13 12:52:47 2006 Subject: festival-freebsoft-utils 0.3 released Message-ID: <87vfh6u51p.fsf@zamazal.org> festival-freebsoft-utils 0.3 released ===================================== The Free(b)soft project is pleased to announce festival-freebsoft-utils 0.3. * What is festival-freebsoft-utils? festival-freebsoft-utils is a collection of Festival utilities that enhance Festival with some useful features. They provide all what is needed for interaction with Speech Dispatcher. Key festival-freebsoft-utils features are: - Generalized concept of input events. festival-freebsoft-utils allows not only plain text synthesis, but also combining it with sounds. Additionally, mechanism of logical events mapped to other events is provided. - Substitution of events for given words. - High-level voice selection mechanism and setting of basic prosodic parameters. - Spelling mode. - Capital letter signalization. - Punctuation modes, for explicit reading or not reading punctuation characters. - Incremental synthesis of texts and events. - Speech Dispatcher support. - Rudimentary SSML support. - Enhancing the Festival extension language with some functions commonly used in Lisp. Support for wrapping already defined Festival functions by your own code. - Everything is written in the extension language, no patching of the Festival C++ sources is needed. * What is new in the 0.3 version? - New modules fileio.scm and recode.scm. - The Speech Dispatcher interface was changed to be compatible with Speech Dispatcher 0.5 and is not compatible with older Speech Dispatcher versions. NOTE: If you use festival-freebsoft-utils with Speech Dispatcher, you must install Speech Dispatcher 0.5 release candidate or higher. There was an incompatible change in the Speech Dispatcher Festival interface that prevents the new festival-freebsoft-utils version to fully cooperate with older Speech Dispatcher versions. * Where to get it? You can get the distribution tarball of the released version from http://www.freebsoft.org/pub/projects/festival-freebsoft-utils/festival-freebsoft-utils-0.3.tar.gz . The home page of the project is http://www.freebsoft.org/festival-freebsoft-utils . From hanke@volny.cz Fri Jul 2 12:58:30 2004 From: Hynek Hanke To: speechd-discuss@nongnu.org Date: Sun Aug 13 12:52:47 2006 Subject: Speech Dispatcher 0.5 Release Candidate 1 available Message-ID: <20040702105830.GB20131@volny.cz> Speech Dispatcher 0.5 Release Candidate 1 ========================================= The Brailcom organization is happy to announce the availability of Speech Dispatcher 0.5-rc1 developed as a part of the Free(b)Soft project. This is a Release Candidate version, so we invite everyone to help us testing it before we finally release the 0.5 version. * What is Speech Dispatcher? Speech Dispatcher is a device independent layer for speech synthesis, developed with the goal of making the usage of speech synthesis easier for application programmers. It takes care of most of the tasks necessary to solve in speech enabled applications. What is a very high level GUI library to graphics, Speech Dispatcher is to speech synthesis. The architecture of Speech Dispatcher is based on a proven client/server model. The basic means of client communication with Speech Dispatcher is through a TCP connection using the Speech Synthesis Interface Protocol (SSIP). Key Speech Dispatcher features are: - Message priority model that allows multiple simultaneous connections to Speech Dispatcher from one or more clients and tries to provide the user with the most important messages. - Different output modules that talk to different synthesizers so that the programmer doesn't need to care which particular synthesizer is being used. Currently Festival, Flite, Epos and (non-free) Dectalk software are supported. - Client-based configuration allows users to configure different settings for different clients that connect to Speech Dispatcher. - Simple interface for programs written in C, C++ provided through a shared library, Common Lisp and Guile interface. The Python interface is being worked on. An Elisp library is developed as a sperate project speechd-el. Possibly an interface to any other language can be developed. * What is new in 0.5-rc1? - SSIP now supports SSML (Speech Synthesis Markup Language) messages. - Communication with Festival is highly improved. - Improved DTK Software Synthesis support. NOTE: The communication mode of the Festival output module (SSML mode) is incompatible with the previous festival-freebsoft-utils releases. You need to install festival-freebsoft-utils 0.3 or higher! * Where to get it? You can get the distribution tarball of the released version from http://www.freebsoft.org/pub/projects/speechd/speech-dispatcher-0.5-rc1.tar.gz You also need to get festival-freebsoft-utils in version 3 or newer http://www.freebsoft.org/pub/projects/festival-freebsoft-utils/ We recommend you to fetch the sound icons Speech Dispatcher can use as well. They are available at http://www.freebsoft.org/pub/projects/sound-icons/sound-icons-0.1.tar.gz Corresponding Debian packages will be available at your Debian distribution mirror as soon as the official version 0.5 is released. (We would highly appreciate any help with RPM distribution!) The home page of the project is http://www.freebsoft.org/speechd * How to report bugs? Please report bugs at . For other contact please use Happy synthesizing! From hanke@volny.cz Fri Jul 2 13:51:13 2004 From: Hynek Hanke To: speechd-discuss@nongnu.org Date: Sun Aug 13 12:52:47 2006 Subject: French language with Speech Dispatcher, Speechd-el and DECtalk software 5.0 In-Reply-To: <16584.17785.22678.280137@gargle.gargle.HOWL> References: <16584.17785.22678.280137@gargle.gargle.HOWL> Message-ID: <20040702115113.GA10832@volny.cz> On Thu, Jun 10, 2004 at 01:26:49PM +0200, Gilles Casse wrote: > A new option has been added to select the language (e.g. for spanish "-l es" ). > At the moment, the default language is french :-). > However, I meet an issue when typing french characters under Emacs: > For example, the c cedilla (0xe7) seems to crash the output module. > When I type: > fran?ais Hi Gilles, thanks for your interest and please excuse the delay. We have switched the internal handling of messages inside the modules to SSML (Speech Synthesis Markup Language) and there were various changes in order to do this. I hope the problem is fixed in the last version 0.5-rc1, please could you try it? There might be some problem in recoding the characters also. Please try to add GenericLanguage "fr" "fr" "iso-8859-?" to the configuration file dtk-generic.conf in etc/speech-dispatcher/modules/ and substitute ? for the appropriate number. (You will most probably need to do this anyway). If it still doesn't work, please switch debugging on in dtk-generic.conf, repeat the situation and send me the resulting file. With Regards, Hynek Hanke From nath.ml@free.fr Fri Jul 2 23:54:50 2004 From: Nath To: speechd-discuss@nongnu.org Date: Sun Aug 13 12:52:47 2006 Subject: French language with Speech Dispatcher, Speechd-el and DECtalk software 5.0 In-Reply-To: <20040702115113.GA10832@volny.cz> (Hynek Hanke's message of "Fri, 2 Jul 2004 13:51:13 +0200") References: <16584.17785.22678.280137@gargle.gargle.HOWL> <20040702115113.GA10832@volny.cz> Message-ID: <87y8m22hxx.fsf@spy.home> Hynek Hanke writes: > On Thu, Jun 10, 2004 at 01:26:49PM +0200, Gilles Casse wrote: >> A new option has been added to select the language (e.g. for spanish "-l es" ). >> At the moment, the default language is french :-). >> However, I meet an issue when typing french characters under Emacs: >> For example, the c cedilla (0xe7) seems to crash the output module. >> When I type: >> fran?ais > > Hi Gilles, > > thanks for your interest and please excuse the delay. We have switched the > internal handling of messages inside the modules to SSML (Speech Synthesis > Markup Language) and there were various changes in order to do this. I hope > the problem is fixed in the last version 0.5-rc1, please could you try it? > > There might be some problem in recoding the characters also. Please try to add > GenericLanguage "fr" "fr" "iso-8859-?" to the configuration > file dtk-generic.conf in etc/speech-dispatcher/modules/ and substitute > ? for the appropriate number. (You will most probably need to do this > anyway). > > If it still doesn't work, please switch debugging on in dtk-generic.conf, > repeat the situation and send me the resulting file. > > With Regards, > Hynek Hanke > Hi Hynek ? al first of all, thx for all the great work you made. I will tru the new speech-dispatcher with DECtalk in france tomorrow and let you know the result of course. Best regards, -- Nath From gcasse@oralux.org Sun Jul 4 09:46:40 2004 From: Gilles Casse To: speechd-discuss@nongnu.org Date: Sun Aug 13 12:52:47 2006 Subject: French language with Speech Dispatcher, Speechd-el and DECtalk software 5.0 In-Reply-To: <20040702115113.GA10832@volny.cz> References: <16584.17785.22678.280137@gargle.gargle.HOWL> <20040702115113.GA10832@volny.cz> Message-ID: <16615.46560.520533.491455@gargle.gargle.HOWL> Hello Hynek, Thanks for your answer. Using the previous release, the generic.c file was patched to avoid the issue. It concerned the language->charset parameter of the g_convert_with_fallback function call; the parameter was replaced with e.g. "UTF-8". I will be glad to try the new 0.5-rc1 release, Cheers Gilles -- Oralux http://oralux.org From jschmude@adelphia.net Tue Jul 6 20:21:22 2004 From: Jacob Schmude To: speechd-discuss@nongnu.org Date: Sun Aug 13 12:52:47 2006 Subject: Bug in generic module? Message-ID: Hi guys Just updated to speech dispatcher 0.5rc1 a little while ago and am noticing a bug in the generic module, which I use to enable software dectalk and speech dispatcher communicate. Everytime it says something, I hear: SPEAK at the beginning and: /SPEAK at the end. This behavior does not exhibit itself with the flite or festival modules. Anyone else notice this, or is my system just being the oddball? THX From hanke@volny.cz Wed Jul 7 16:15:18 2004 From: Hynek Hanke To: speechd-discuss@nongnu.org Date: Sun Aug 13 12:52:47 2006 Subject: Bug in generic module? In-Reply-To: References: Message-ID: <20040707141518.GA2478@volny.cz> On Tue, Jul 06, 2004 at 02:21:22PM -0400, Jacob Schmude wrote: > Just updated to speech dispatcher 0.5rc1 a little while ago and am > noticing a bug in the generic module, which I use to enable software > dectalk and speech dispatcher communicate. Everytime it says something, I > hear: > SPEAK Thank you Jacob. I already know about this issue. It's a trivial error in the generic module, I'll shortly fix it. With Regards, Hynek From hanke@volny.cz Tue Jul 13 22:45:38 2004 From: Hynek Hanke To: speechd-discuss@nongnu.org Date: Sun Aug 13 12:52:47 2006 Subject: French language with Speech Dispatcher, Speechd-el and DECtalk software 5.0 In-Reply-To: <87y8m22hxx.fsf@spy.home> References: <16584.17785.22678.280137@gargle.gargle.HOWL> <20040702115113.GA10832@volny.cz> <87y8m22hxx.fsf@spy.home> Message-ID: <20040713204538.GA1616@volny.cz> On Fri, Jul 02, 2004 at 11:54:50PM +0200, Nath wrote: > I will tru the new speech-dispatcher > with DECtalk in france tomorrow and let you know the result of course. Hi Nath, there is a new tarball (release candidate) on http://www.freebsoft.org/pub/projects/speechd/speech-dispatcher-0.5-rc2.tar.gz Could you please look at if it fixes your problems with french DecTalk you described me on LSM? Be sure to save your dtk-generic.conf somewhere and then move it back, since ``make install'' overwrites it by default and your say_fr is a bit different from the default DecTalk's say, so it needs the modified conf file. You can even backup and restore the whole etc/speech-dispatcher . With Regards, Hynek From gcasse@oralux.org Wed Jul 14 00:57:30 2004 From: Gilles Casse To: speechd-discuss@nongnu.org Date: Sun Aug 13 12:52:47 2006 Subject: French language with Speech Dispatcher, Speechd-el and DECtalk software 5.0 In-Reply-To: <20040702115113.GA10832@volny.cz> References: <16584.17785.22678.280137@gargle.gargle.HOWL> <20040702115113.GA10832@volny.cz> Message-ID: <16628.26842.992087.399033@gargle.gargle.HOWL> Hello Hynek, I was writing a mail about 0.5 rc1 when I saw that rc2 is now available :-). So, the 0.5 rc2 release has been installed to test the french characters using the dtk-generic module for DECtalk 5.0. I am using emacs in console mode and iso 8859-1. When the c cedilla is typed under emacs, the dtk-generic processes crash. A sample from debug-dtk-generic is copied at the end of this post [1]. In generic.c, g_convert_with_fallback expects a UTF-8 character but a single byte, e.g the c cedilla character, is received. Then g_convert_with_fallback returns NULL. I have added a few lines for the test in generic.c which also avoid the crash [2]. I am aware that this is not a solution. The related debug-dtk-generic file using thes lines is also copied at the end [3]. Cheers, Gilles *** [1] Before any patch: the n letter is typed, then the c cedilla character: Tue Jul 13 22:12:08 2004 [603330]: Waiting for child... Tue Jul 13 22:12:08 2004 [603437]: child terminated -: status:0 signal?:1 signal number:9. Tue Jul 13 22:12:08 2004 [604020]: speak() Tue Jul 13 22:12:08 2004 [604112]: Recoding from UTF-8 to iso-8859-1... Tue Jul 13 22:12:08 2004 [604251]: in stripping ssml: |a| Tue Jul 13 22:12:08 2004 [604340]: Requested data: |a| Tue Jul 13 22:12:08 2004 [604463]: Semaphore on Tue Jul 13 22:12:08 2004 [605114]: Entering parent process, closing pipes Tue Jul 13 22:12:08 2004 [605289]: Looping... Tue Jul 13 22:12:08 2004 [605380]: Returned 1 bytes from get_part Tue Jul 13 22:12:08 2004 [605466]: Sending buf to child:|a| 1 Tue Jul 13 22:12:08 2004 [605553]: going to write 1 bytes Tue Jul 13 22:12:08 2004 [605643]: written 1 bytes Tue Jul 13 22:12:08 2004 [605728]: Waiting for response from child... Tue Jul 13 22:12:08 2004 [605855]: Generic: leaving write() normaly Tue Jul 13 22:12:08 2004 [606581]: Starting child... Tue Jul 13 22:12:08 2004 [606688]: UnBlocking user signal Tue Jul 13 22:12:08 2004 [606806]: Entering child loop Tue Jul 13 22:12:08 2004 [606900]: read 1 bytes in child Tue Jul 13 22:12:08 2004 [606986]: text read is: |a| Tue Jul 13 22:12:08 2004 [607111]: child: escaped text is |a| Tue Jul 13 22:12:08 2004 [607236]: child: synth command = |echo "[:np][:ra 338][:dv ap 225]" >/tmp/dtk-speak.txt && echo "a" | fmt >>/tmp/dtk-speak.txt && say -fi /tmp/dtk-speak.txt| Tue Jul 13 22:12:08 2004 [607325]: Speaking in child... Tue Jul 13 22:12:08 2004 [607409]: Blocking user signal Tue Jul 13 22:12:09 2004 [109097]: generic: stop() Tue Jul 13 22:12:09 2004 [109225]: generic: stopping process group pid 23488 Tue Jul 13 22:12:09 2004 [138564]: generic: stop() Tue Jul 13 22:12:09 2004 [138693]: generic: stopping process group pid 23488 Tue Jul 13 22:12:09 2004 [140016]: parent: Read bytes 0, child stopped Tue Jul 13 22:12:09 2004 [140117]: End of data in parent, closing pipes Tue Jul 13 22:12:09 2004 [140224]: Waiting for child... Tue Jul 13 22:12:09 2004 [140330]: child terminated -: status:0 signal?:1 signal number:9. Tue Jul 13 22:12:09 2004 [140883]: speak() *** [2] In generic.c, function module_speak, around line 100, a few lines have been added for the test (at the end of the following sample) if (generic_msg_language->charset != NULL){ DBG("Recoding from UTF-8 to %s...", generic_msg_language->charset); tmp = (char*) g_convert_with_fallback(data, bytes, generic_msg_language->charset, "UTF-8", GenericRecodeFallback, NULL, NULL, NULL); }else{ DBG("Warning: Recoding to language default iso, might not be ok..."); tmp = module_recode_to_iso(data, bytes, generic_msg_language->name, GenericRecodeFallback); } //// (GC) Added for the test if (tmp==NULL) { DBG("tmp=null; data=%s, bytes=%d\n", data, bytes); *generic_message = module_strip_ssml(data); } else { //// // Original code *generic_message = module_strip_ssml(tmp); xfree(tmp); // } module_strip_punctuation_some(*generic_message, GenericStripPunctChars); *** [3] The n letter is typed, then the c cedilla character, the a charcater Tue Jul 13 23:06:02 2004 [117713]: generic: stop() Tue Jul 13 23:06:02 2004 [117842]: generic: stopping process group pid 30909 Tue Jul 13 23:06:02 2004 [120428]: parent: Read bytes 0, child stopped Tue Jul 13 23:06:02 2004 [120530]: End of data in parent, closing pipes Tue Jul 13 23:06:02 2004 [120637]: Waiting for child... Tue Jul 13 23:06:02 2004 [120747]: child terminated -: status:0 signal?:1 signal number:9. Tue Jul 13 23:06:02 2004 [147927]: speak() Tue Jul 13 23:06:02 2004 [148049]: Recoding from UTF-8 to iso-8859-1... Tue Jul 13 23:06:02 2004 [148200]: in stripping ssml: |n| Tue Jul 13 23:06:02 2004 [148290]: Requested data: |n| Tue Jul 13 23:06:02 2004 [148425]: Semaphore on Tue Jul 13 23:06:02 2004 [149094]: Entering parent process, closing pipes Tue Jul 13 23:06:02 2004 [149269]: Looping... Tue Jul 13 23:06:02 2004 [149359]: Returned 1 bytes from get_part Tue Jul 13 23:06:02 2004 [149445]: Sending buf to child:|n| 1 Tue Jul 13 23:06:02 2004 [149532]: going to write 1 bytes Tue Jul 13 23:06:02 2004 [149622]: written 1 bytes Tue Jul 13 23:06:02 2004 [149708]: Waiting for response from child... Tue Jul 13 23:06:02 2004 [149833]: Generic: leaving write() normaly Tue Jul 13 23:06:02 2004 [150520]: Starting child... Tue Jul 13 23:06:02 2004 [150624]: UnBlocking user signal Tue Jul 13 23:06:02 2004 [150742]: Entering child loop Tue Jul 13 23:06:02 2004 [150961]: read 1 bytes in child Tue Jul 13 23:06:02 2004 [151050]: text read is: |n| Tue Jul 13 23:06:02 2004 [151175]: child: escaped text is |n| Tue Jul 13 23:06:02 2004 [151301]: child: synth command = |echo "[:np][:ra 338][:dv ap 225]" >/tmp/dtk-speak.txt && echo "n" | fmt >>/tmp/dtk-speak.txt && say -fi /tmp/dtk-speak.txt| Tue Jul 13 23:06:02 2004 [151390]: Speaking in child... Tue Jul 13 23:06:02 2004 [151474]: Blocking user signal Tue Jul 13 23:06:03 2004 [101965]: generic: stop() Tue Jul 13 23:06:03 2004 [102092]: generic: stopping process group pid 30921 Tue Jul 13 23:06:03 2004 [104916]: parent: Read bytes 0, child stopped Tue Jul 13 23:06:03 2004 [105020]: End of data in parent, closing pipes Tue Jul 13 23:06:03 2004 [105127]: Waiting for child... Tue Jul 13 23:06:03 2004 [105236]: child terminated -: status:0 signal?:1 signal number:9. Tue Jul 13 23:06:03 2004 [132338]: speak() Tue Jul 13 23:06:03 2004 [132458]: Recoding from UTF-8 to iso-8859-1... Tue Jul 13 23:06:03 2004 [132810]: tmp=null; data=?, bytes=1 Tue Jul 13 23:06:03 2004 [132905]: in stripping ssml: |?| Tue Jul 13 23:06:03 2004 [132992]: Requested data: |?| Tue Jul 13 23:06:03 2004 [133128]: Semaphore on Tue Jul 13 23:06:03 2004 [133796]: Entering parent process, closing pipes Tue Jul 13 23:06:03 2004 [133942]: Looping... Tue Jul 13 23:06:03 2004 [134029]: Returned 1 bytes from get_part Tue Jul 13 23:06:03 2004 [134114]: Sending buf to child:|?| 1 Tue Jul 13 23:06:03 2004 [134201]: going to write 1 bytes Tue Jul 13 23:06:03 2004 [134292]: written 1 bytes Tue Jul 13 23:06:03 2004 [134378]: Waiting for response from child... Tue Jul 13 23:06:03 2004 [134502]: Generic: leaving write() normaly Tue Jul 13 23:06:03 2004 [135165]: Starting child... Tue Jul 13 23:06:03 2004 [135271]: UnBlocking user signal Tue Jul 13 23:06:03 2004 [135389]: Entering child loop Tue Jul 13 23:06:03 2004 [135484]: read 1 bytes in child Tue Jul 13 23:06:03 2004 [135570]: text read is: |?| Tue Jul 13 23:06:03 2004 [135735]: child: escaped text is |?| Tue Jul 13 23:06:03 2004 [135865]: child: synth command = |echo "[:np][:ra 338][:dv ap 225]" >/tmp/dtk-speak.txt && echo "?" | fmt >>/tmp/dtk-speak.txt && say -fi /tmp/dtk-speak.txt| Tue Jul 13 23:06:03 2004 [135954]: Speaking in child... Tue Jul 13 23:06:03 2004 [136038]: Blocking user signal Tue Jul 13 23:06:03 2004 [984775]: Executed shell command returned with 0 Tue Jul 13 23:06:03 2004 [984920]: UnBlocking user signal Tue Jul 13 23:06:03 2004 [985037]: child->parent: ok, send more data Tue Jul 13 23:06:03 2004 [985166]: Ok, received report to continue... Tue Jul 13 23:06:03 2004 [985260]: Looping... Tue Jul 13 23:06:03 2004 [985345]: Returned -1 bytes from get_part Tue Jul 13 23:06:03 2004 [985430]: End of data in parent, closing pipes Tue Jul 13 23:06:03 2004 [985532]: read 0 bytes in child Tue Jul 13 23:06:03 2004 [985623]: child: Pipe closed, exiting, closing pipes.. Tue Jul 13 23:06:03 2004 [985724]: Child ended... Tue Jul 13 23:06:03 2004 [986518]: Waiting for child... Tue Jul 13 23:06:03 2004 [986633]: child terminated -: status:1 signal?:0 signal number:0. Tue Jul 13 23:06:04 2004 [557069]: speak() Tue Jul 13 23:06:04 2004 [557201]: Recoding from UTF-8 to iso-8859-1... Tue Jul 13 23:06:04 2004 [557347]: in stripping ssml: |a| Tue Jul 13 23:06:04 2004 [557438]: Requested data: |a| Tue Jul 13 23:06:04 2004 [557572]: Semaphore on Tue Jul 13 23:06:04 2004 [558235]: Entering parent process, closing pipes Tue Jul 13 23:06:04 2004 [558378]: Looping... Tue Jul 13 23:06:04 2004 [558464]: Returned 1 bytes from get_part Tue Jul 13 23:06:04 2004 [558551]: Sending buf to child:|a| 1 Tue Jul 13 23:06:04 2004 [558638]: going to write 1 bytes Tue Jul 13 23:06:04 2004 [558728]: written 1 bytes Tue Jul 13 23:06:04 2004 [558814]: Waiting for response from child... Tue Jul 13 23:06:04 2004 [558938]: Generic: leaving write() normaly Tue Jul 13 23:06:04 2004 [588799]: Starting child... Tue Jul 13 23:06:04 2004 [589042]: UnBlocking user signal Tue Jul 13 23:06:04 2004 [589164]: Entering child loop Tue Jul 13 23:06:04 2004 [589262]: read 1 bytes in child Tue Jul 13 23:06:04 2004 [589348]: text read is: |a| Tue Jul 13 23:06:04 2004 [589521]: child: escaped text is |a| Tue Jul 13 23:06:04 2004 [589650]: child: synth command = |echo "[:np][:ra 338][:dv ap 225]" >/tmp/dtk-speak.txt && echo "a" | fmt >>/tmp/dtk-speak.txt && say -fi /tmp/dtk-speak.txt| Tue Jul 13 23:06:04 2004 [589739]: Speaking in child... Tue Jul 13 23:06:04 2004 [589823]: Blocking user signal Tue Jul 13 23:06:05 2004 [744758]: Executed shell command returned with 0 Tue Jul 13 23:06:05 2004 [744904]: UnBlocking user signal Tue Jul 13 23:06:05 2004 [745020]: child->parent: ok, send more data Tue Jul 13 23:06:05 2004 [745151]: Ok, received report to continue... Tue Jul 13 23:06:05 2004 [745245]: Looping... Tue Jul 13 23:06:05 2004 [745331]: Returned -1 bytes from get_part Tue Jul 13 23:06:05 2004 [745417]: End of data in parent, closing pipes Tue Jul 13 23:06:05 2004 [745519]: read 0 bytes in child Tue Jul 13 23:06:05 2004 [745610]: child: Pipe closed, exiting, closing pipes.. Tue Jul 13 23:06:05 2004 [745712]: Child ended... Tue Jul 13 23:06:05 2004 [746502]: Waiting for child... Tue Jul 13 23:06:05 2004 [746618]: child terminated -: status:1 signal?:0 signal number:0. -- Oralux http://oralux.org From hanke@volny.cz Sun Jul 18 18:11:53 2004 From: Hynek Hanke To: speechd-discuss@nongnu.org Date: Sun Aug 13 12:52:47 2006 Subject: French language with Speech Dispatcher, Speechd-el and DECtalk software 5.0 In-Reply-To: <16628.26842.992087.399033@gargle.gargle.HOWL> References: <16584.17785.22678.280137@gargle.gargle.HOWL> <20040702115113.GA10832@volny.cz> <16628.26842.992087.399033@gargle.gargle.HOWL> Message-ID: <20040718161153.GA2467@volny.cz> On Wed, Jul 14, 2004 at 12:57:30AM +0200, Gilles Casse wrote: > When the c cedilla is typed under emacs, the dtk-generic processes crash. > In generic.c, g_convert_with_fallback expects a UTF-8 character but a single byte, > e.g the c cedilla character, is received. > Then g_convert_with_fallback returns NULL. Yes, now I see it. The same happens in Czech language, speechd-el doesn't send the characters (at least with the KEY command) in UTF-8 in SSIP, but in the original charset. This problem doesn't appear with longer messages sent through the SPEAK command, these are correctly in UTF-8. Could you Milan look at it please? I'm not familiar with speechd-el enough. > I have added a few lines for the test in generic.c which also avoid the crash [2]. Thank you for spotting this, there needs to be a safe test for this case. With Regards, Hynek Hanke From pdm@freebsoft.org Sun Jul 18 22:49:39 2004 From: Milan Zamazal To: speechd-discuss@nongnu.org Date: Sun Aug 13 12:52:47 2006 Subject: French language with Speech Dispatcher, Speechd-el and DECtalk software 5.0 In-Reply-To: <20040718161153.GA2467@volny.cz> (Hynek Hanke's message of "Sun, 18 Jul 2004 18:11:53 +0200") References: <16584.17785.22678.280137@gargle.gargle.HOWL> <20040702115113.GA10832@volny.cz> <16628.26842.992087.399033@gargle.gargle.HOWL> <20040718161153.GA2467@volny.cz> Message-ID: <87658lyra4.fsf@swan.zamazal.org> >>>>> "HH" == Hynek Hanke writes: HH> speechd-el doesn't send the characters (at least with the KEY HH> command) in UTF-8 in SSIP, but in the original charset. This HH> problem doesn't appear with longer messages sent through the HH> SPEAK command, these are correctly in UTF-8. Could you Milan HH> look at it please? Any string sent to SSIP is processed through the same routine, performing the conversion to UTF-8, in speechd-el. So if the encoding differs within the same connection, something is weird. I'm sure there are UTF-8 related bugs in Emacs 21.3, but I can't reproduce the particular problem you describe. Are you sure the wrong coding comes from speechd-el? If so, what does the call (encode-coding-string "KEY ?\n" speechd--coding-system) return? And what's the value of speechd--coding-system? Regards, Milan Zamazal -- Free software is about freedom, not about free beer. If you care only about the latter, you'll end up with no freedom and no free beer. From hanke@volny.cz Sun Jul 18 23:31:04 2004 From: Hynek Hanke To: speechd-discuss@nongnu.org Date: Sun Aug 13 12:52:47 2006 Subject: French language with Speech Dispatcher, Speechd-el and DECtalk software 5.0 In-Reply-To: <87658lyra4.fsf@swan.zamazal.org> References: <16584.17785.22678.280137@gargle.gargle.HOWL> <20040702115113.GA10832@volny.cz> <16628.26842.992087.399033@gargle.gargle.HOWL> <20040718161153.GA2467@volny.cz> <87658lyra4.fsf@swan.zamazal.org> Message-ID: <20040718213104.GB3560@volny.cz> On Sun, Jul 18, 2004 at 10:49:39PM +0200, Milan Zamazal wrote: > Any string sent to SSIP is processed through the same routine, > performing the conversion to UTF-8, in speechd-el. So if the encoding > differs within the same connection, something is weird. I'm sure there > are UTF-8 related bugs in Emacs 21.3, but I can't reproduce the > particular problem you describe. You probably can't reproduce it because it was fixed in the latest CVS releases of speechd-el. I'm not sure about debian packages. After updating my CVS tree, the c cedilla works perfectly. Can Gilles confirm it also works with Dectalk? Thank you, Hynek From pdm@freebsoft.org Mon Jul 19 08:14:47 2004 From: Milan Zamazal To: speechd-discuss@nongnu.org Date: Sun Aug 13 12:52:47 2006 Subject: French language with Speech Dispatcher, Speechd-el and DECtalk software 5.0 In-Reply-To: <20040718213104.GB3560@volny.cz> (Hynek Hanke's message of "Sun, 18 Jul 2004 23:31:04 +0200") References: <16584.17785.22678.280137@gargle.gargle.HOWL> <20040702115113.GA10832@volny.cz> <16628.26842.992087.399033@gargle.gargle.HOWL> <20040718161153.GA2467@volny.cz> <87658lyra4.fsf@swan.zamazal.org> <20040718213104.GB3560@volny.cz> Message-ID: <87acxwr0a0.fsf@swan.zamazal.org> >>>>> "HH" == Hynek Hanke writes: HH> You probably can't reproduce it because it was fixed in the HH> latest CVS releases of speechd-el. No, I tried it with speechd-el 0.4 and it worked fine. HH> After updating my CVS tree, the c cedilla works perfectly. I couldn't find in the change log since the 0.4 release any change which was likely to be related to this problem. So beware, the bug is probably elsewhere (Emacs UTF-8 bug still possible). Regards, Milan Zamazal -- real programmer? don't get me started. if you need to hide your pathetic excuse for a carreer behind super-macho languages like C, C++, and/or Perl instead of writing clean, maintainable, efficient code, you aren't much of a real programmer in my view. -- Erik Naggum in comp.emacs From gcasse@oralux.org Mon Jul 19 22:21:02 2004 From: Gilles Casse To: speechd-discuss@nongnu.org Date: Sun Aug 13 12:52:47 2006 Subject: French language with Speech Dispatcher, Speechd-el and DECtalk software 5.0 In-Reply-To: <20040718213104.GB3560@volny.cz> References: <16584.17785.22678.280137@gargle.gargle.HOWL> <20040702115113.GA10832@volny.cz> <16628.26842.992087.399033@gargle.gargle.HOWL> <20040718161153.GA2467@volny.cz> <87658lyra4.fsf@swan.zamazal.org> <20040718213104.GB3560@volny.cz> Message-ID: <16636.11566.659181.697548@gargle.gargle.HOWL> Hello all, Using the last CVS release of speechd-el, the c cedilla worked but after a few tries, an e acute caused again the issue. The test gives this result here: (encode-coding-string "KEY ?\n" speechd--coding-system) "KEY ??? " In hexa: 22 4b 45 59 20 c3 c2 a1 0d 0a 22 0a " K E Y ? ? ? \r \n " \n speechd--coding-system's value is utf-8-dos Regards, Gilles -- Oralux http://oralux.org From pdm@freebsoft.org Tue Jul 20 11:00:46 2004 From: Milan Zamazal To: speechd-discuss@nongnu.org Date: Sun Aug 13 12:52:47 2006 Subject: French language with Speech Dispatcher, Speechd-el and DECtalk software 5.0 In-Reply-To: <16636.11566.659181.697548@gargle.gargle.HOWL> (Gilles Casse's message of "Mon, 19 Jul 2004 22:21:02 +0200") References: <16584.17785.22678.280137@gargle.gargle.HOWL> <20040702115113.GA10832@volny.cz> <16628.26842.992087.399033@gargle.gargle.HOWL> <20040718161153.GA2467@volny.cz> <87658lyra4.fsf@swan.zamazal.org> <20040718213104.GB3560@volny.cz> <16636.11566.659181.697548@gargle.gargle.HOWL> Message-ID: <87n01vt5mp.fsf@blackbird.wifi.zamazal.org> >>>>> "GC" == Gilles Casse writes: GC> Using the last CVS release of speechd-el, the c cedilla worked GC> but after a few tries, an e acute caused again the issue. So you mean the problem is irregular and it sometimes occurs and sometimes not? GC> The test gives this result here: GC> (encode-coding-string "KEY ?\n" speechd--coding-system) GC> "KEY ??? " GC> In hexa: GC> 22 4b 45 59 20 c3 c2 a1 0d 0a 22 0a GC> " K E Y ? ? ? \r \n " \n ^^^ GC> speechd--coding-system's value is utf-8-dos The #c2 character is an extra character, it shouldn't be there. Other characters are correct. But it doesn't mean the #c2 character is actually in the SSIP output -- it may just be added during conversions between different Emacs buffer encodings. Could you please do the following additional testing so that we are more sure what happens? - Enable protocol logging in Speech Dispatcher by uncommenting and setting the CustomLogFile option in speechd.conf, e.g. CustomLogFile "protocol" "/var/log/speech-dispatcher/speech-dispatcher-protocol.log" - Load the following Elisp function: (defun test-utf8 (string) (let ((present (lambda (string) (let ((i 0) (len (length string)) (result (list ': (string-bytes string) string))) (while (< i len) (setq result (cons (aref string i) result)) (setq i (1+ i))) (reverse result))))) (list (funcall present string) (funcall present (encode-coding-string string 'utf-8-dos))))) - When the problem occurs, look into the protocol log in the file defined above and check whether the corresponding KEY command is encoded correctly there or not. Please make sure that what you see/check is actually present in the log file, i.e. that you are not confused by a non-UTF buffer or terminal encoding. - Eval the call (test-utf8 "KEY ?\n") Try to substite both the working and non-working keys for `?' and send the results. Thanks for testing. Regards, Milan Zamazal -- Wasting somebody else's time strikes me as the height of rudeness. Bill Gates From gcasse@oralux.org Wed Jul 21 00:24:02 2004 From: Gilles Casse To: speechd-discuss@nongnu.org Date: Sun Aug 13 12:52:47 2006 Subject: French language with Speech Dispatcher, Speechd-el and DECtalk software 5.0 In-Reply-To: <87n01vt5mp.fsf@blackbird.wifi.zamazal.org> References: <16584.17785.22678.280137@gargle.gargle.HOWL> <20040702115113.GA10832@volny.cz> <16628.26842.992087.399033@gargle.gargle.HOWL> <20040718161153.GA2467@volny.cz> <87658lyra4.fsf@swan.zamazal.org> <20040718213104.GB3560@volny.cz> <16636.11566.659181.697548@gargle.gargle.HOWL> <87n01vt5mp.fsf@blackbird.wifi.zamazal.org> Message-ID: <16637.39810.539320.476991@gargle.gargle.HOWL> Hi Milan, The test conditions were probably different yesterday since today, typing the c cedilla right away stops the dtk-generic processes. A fragment of the protocol log file [1] is copied at the end. At the very beginning, there is the single byte e7 for the c cedilla: 12:DATA:|KEY ? and around the end, its utf-8 value is displayed: 12:DATA:|KEY ?? The call to test-utf8 returns: (test-utf8 "?") returns: (("?" 2 : 2279) ("??" 2 : 195 167)) 195 167 matches the utf-8 value of c cedilla Regards, Gilles [1] protocol log file [Tue Jul 20 22:14:59 2004] speechd: 12:DATA:|BLOCK BEGIN | [Tue Jul 20 22:14:59 2004] speechd: 12:REPLY:|260 OK INSIDE BLOCK | [Tue Jul 20 22:14:59 2004] speechd: 12:DATA:|KEY ? | [Tue Jul 20 22:14:59 2004] speechd: 12:REPLY:|225 OK MESSAGE QUEUED | [Tue Jul 20 22:14:59 2004] speechd: Command sent to output module: |SET | [Tue Jul 20 22:14:59 2004] speechd: Reply from output module: |203 OK RECEIVING SETTINGS US SENT | [Tue Jul 20 22:14:59 2004] speechd: Command sent to output module: |pitch=0 rate=0 volume=100 punctuation_mode=some spelling_mode=off cap_let_recogn=none language=en voice=male1 | [Tue Jul 20 22:14:59 2004] speechd: Command sent to output module: |. | [Tue Jul 20 22:14:59 2004] speechd: Reply from output module: |203 OK SETTINGS RECEIVED US SENT | [Tue Jul 20 22:14:59 2004] speechd: Command sent to output module: |KEY | [Tue Jul 20 22:14:59 2004] speechd: Reply from output module: |202 OK RECEIVING MESSAGE US SENT | [Tue Jul 20 22:14:59 2004] speechd: Command sent to output module: |?| [Tue Jul 20 22:14:59 2004] speechd: Command sent to output module: | . | [Tue Jul 20 22:14:59 2004] speechd: 12:DATA:|BLOCK END | [Tue Jul 20 22:14:59 2004] speechd: 12:REPLY:|261 OK OUTSIDE BLOCK | [Tue Jul 20 22:14:59 2004] speechd: 12:DATA:|SET self PRIORITY TEXT | [Tue Jul 20 22:14:59 2004] speechd: 12:REPLY:|202 OK PRIORITY SET | [Tue Jul 20 22:14:59 2004] speechd: 12:DATA:|CANCEL self | [Tue Jul 20 22:15:01 2004] speechd: 12:REPLY:|213 OK CANCELED | [Tue Jul 20 22:15:01 2004] speechd: 12:DATA:|SET self PRIORITY NOTIFICATION | [Tue Jul 20 22:15:01 2004] speechd: 12:REPLY:|202 OK PRIORITY SET | [Tue Jul 20 22:15:01 2004] speechd: 12:DATA:|BLOCK BEGIN | [Tue Jul 20 22:15:01 2004] speechd: 12:REPLY:|260 OK INSIDE BLOCK | [Tue Jul 20 22:15:01 2004] speechd: 12:DATA:|KEY ?? | [Tue Jul 20 22:15:01 2004] speechd: 12:REPLY:|225 OK MESSAGE QUEUED | [Tue Jul 20 22:15:01 2004] speechd: 12:DATA:|BLOCK END | [Tue Jul 20 22:15:01 2004] speechd: 12:REPLY:|261 OK OUTSIDE BLOCK | [Tue Jul 20 22:15:01 2004] speechd: 12:DATA:|SET self PRIORITY TEXT | [Tue Jul 20 22:15:01 2004] speechd: 12:REPLY:|202 OK PRIORITY SET | [Tue Jul 20 22:15:01 2004] speechd: 12:DATA:|CANCEL self | [Tue Jul 20 22:15:01 2004] speechd: 12:REPLY:|213 OK CANCELED | [Tue Jul 20 22:15:01 2004] speechd: 12:DATA:|SET self PRIORITY NOTIFICATION | -- Oralux http://oralux.org From pdm@freebsoft.org Wed Jul 21 21:04:58 2004 From: Milan Zamazal To: speechd-discuss@nongnu.org Date: Sun Aug 13 12:52:47 2006 Subject: French language with Speech Dispatcher, Speechd-el and DECtalk software 5.0 In-Reply-To: <16637.39810.539320.476991@gargle.gargle.HOWL> (Gilles Casse's message of "Wed, 21 Jul 2004 00:24:02 +0200") References: <16584.17785.22678.280137@gargle.gargle.HOWL> <20040702115113.GA10832@volny.cz> <16628.26842.992087.399033@gargle.gargle.HOWL> <20040718161153.GA2467@volny.cz> <87658lyra4.fsf@swan.zamazal.org> <20040718213104.GB3560@volny.cz> <16636.11566.659181.697548@gargle.gargle.HOWL> <87n01vt5mp.fsf@blackbird.wifi.zamazal.org> <16637.39810.539320.476991@gargle.gargle.HOWL> Message-ID: <871xj5uqp1.fsf@zamazal.org> >>>>> "GC" == Gilles Casse writes: GC> The test conditions were probably different yesterday since GC> today, typing the c cedilla right away stops the dtk-generic GC> processes. GC> A fragment of the protocol log file [1] is copied at the end. GC> At the very beginning, there is the single byte e7 for the c GC> cedilla: 12:DATA:|KEY ? GC> and around the end, its utf-8 value is displayed: 12:DATA:|KEY GC> ?? Strange, strange. The only explanation I can think of is that it's really an Emacs bug. Is there anyone else here, who can reproduce the problem? I'm still unable to reproduce it. We can try to the following now: - Try to replace the speechd.el file with the following version: -------------- next part -------------- A non-text attachment was scrubbed... Name: speechd.el Type: application/octet-stream Size: 43138 bytes Desc: not available Url : http://lists.freebsoft.org/pipermail/speechd/attachments/20040721/6e3d54d9/speechd.obj -------------- next part -------------- It tries to catch and replace given SSIP commands just before sending them to the socket. See the variable speechd-utf-8-hack, you can alter its value and perform various experiments with it. Please note this speechd.el file is encoded in Latin-1, as is defined in the local variables at its end. - I think it would be worth to try to run speechd-el with the Emacs CVS snapshot, to see whether the bug is present there too. speechd-el runs fine with the Emacs CVS version (and the .elc files are compatible too), so all you need to do is to compile the Emacs snapshot. See http://savannah.gnu.org/cvs/?group=emacs for instructions how to get Emacs from its CVS archive. Regards, Milan Zamazal -- It is the reformer, who is anxious for the reform, and not society, from which he should expect nothing better than opposition, abhorrence and even mortal persecution. -- M. K. Gandhi From pdm@freebsoft.org Thu Jul 22 11:03:17 2004 From: Milan Zamazal To: speechd-discuss@nongnu.org Date: Sun Aug 13 12:52:47 2006 Subject: Where to report bugs on Free(b)soft packages to Message-ID: <87658gtnvu.fsf@zamazal.org> When reporting bugs on Free(b)soft packages, please use the appropriate bug tracking e-mail addresses. Each of the packages has its own address of the form PACKAGE@bugs.freebsoft.org, where PACKAGE is to be replaced by the package name. Namely: festival-czech -> festival-czech@bugs.freebsoft.org festival-freebsoft-utils -> festival-freebsoft-utils@bugs.freebsoft.org Speech Dispatcher -> speech-dispatcher@bugs.freebsoft.org or speechd@bugs.freebsoft.org speechd-el -> speechd-el@bugs.freebsoft.org bug tracking software -> sbts@bugs.freebsoft.org If you are not sure which package the bug is in, send it on the package you think it is in. If you have no idea which package to send the bug report to or there is no bug reporting address for a particular package, use the address unknown@bugs.freebsoft.org Operational requests on the bug tracking system, like requests for new categories etc., can be sent to the address bts@bugs.freebsoft.org If you like to play with the bug tracking system or to test it, use the address test@bugs.freebsoft.org If you are completely confused :-), send the bug report wherever we are likely to read it. Most important is to actually report the bug. Thank you. Regards, Milan Zamazal -- Wasting somebody else's time strikes me as the height of rudeness. Bill Gates From hanke@volny.cz Mon Oct 18 23:05:10 2004 From: Hynek Hanke To: speechd-discuss@nongnu.org Date: Sun Aug 13 12:52:47 2006 Subject: Speech Dispatcher 0.5 released Message-ID: <20041018210510.GD2220@volny.cz> Speech Dispatcher 0.5 ===================== The Brailcom organization is happy to announce the availability of Speech Dispatcher 0.5 developed as a part of the Free(b)Soft project. * What is Speech Dispatcher? Speech Dispatcher is a device independent layer for speech synthesis, developed with the goal of making the usage of speech synthesis easier for application programmers. It takes care of most of the tasks necessary to solve in speech enabled applications. What is a very high level GUI library to graphics, Speech Dispatcher is to speech synthesis. The architecture of Speech Dispatcher is based on a proven client/server model. The basic means of client communication with Speech Dispatcher is through a TCP connection using the Speech Synthesis Interface Protocol (SSIP). Key Speech Dispatcher features are: - Message priority model that allows multiple simultaneous connections to Speech Dispatcher from one or more clients and tries to provide the user with the most important messages. - Different output modules that talk to different synthesizers so that the programmer doesn't need to care which particular synthesizer is being used. Currently Festival, Flite, Epos and (non-free) Dectalk software are supported. - Client-based configuration allows users to configure different settings for different clients that connect to Speech Dispatcher. - Simple interface for programs written in C, C++ provided through a shared library, Common Lisp and Guile interface. Full Python interface is newly available. An Elisp library is developed as a sperate project speechd-el. Possibly an interface to any other language can be developed. * What is new in 0.5? - SSIP now supports SSML (Speech Synthesis Markup Language) messages. - Communication with Festival is highly improved. - Improved DTK Software Synthesis support. - Python interface library now supports full SSIP. - Various bugfixes. NOTE: The communication mode of the Festival output module (SSML mode) is incompatible with the previous festival-freebsoft-utils releases. You need to install festival-freebsoft-utils 0.3 or higher to be able to run Speech Dispatcher with Festival! * Where to get it? You can get the distribution tarball of the released version from http://www.freebsoft.org/pub/projects/speechd/speech-dispatcher-0.5.tar.gz You also need to get festival-freebsoft-utils in version 0.3 or newer http://www.freebsoft.org/pub/projects/festival-freebsoft-utils/ We recommend you to fetch the sound icons Speech Dispatcher can use as well. They are available at http://www.freebsoft.org/pub/projects/sound-icons/sound-icons-0.1.tar.gz Corresponding Debian packages will be shortly available at your Debian distribution mirror. (We would highly appreciate any help with RPM distribution!) The home page of the project is http://www.freebsoft.org/speechd * How to report bugs? Please report bugs at . For other contact please use Happy synthesizing! From pdm@freebsoft.org Tue Oct 19 15:53:45 2004 From: Milan Zamazal To: speechd-discuss@nongnu.org Date: Sun Aug 13 12:52:47 2006 Subject: speechd-el 0.5 released Message-ID: <87d5zex192.fsf@blackbird.zamazal.org> speechd-el 0.5 released ======================= The Free(b)soft project is pleased to announce speechd-el 0.5. * What is speechd-el? speechd-el is an Emacs client to Speech Dispatcher, providing a complex speech interface to Emacs, focused especially on (but not limited to) the blind and visually impaired users. It allows the user to work with Emacs without looking on the screen, using the speech output produced by the synthesizers supported in Speech Dispatcher. Key speechd-el features are: - Automated speaking of most Emacs messages and actions, without necessity to customize each particular Emacs function to make it speak. - Most built-in and external Emacs packages can produce speech output immediately, without any special support. - Almost no change of the standard Emacs behavior, no interference with user settings. - Highly configurable. - Full cooperation with Speech Dispatcher. - Small code size. - Free Software, distributed under the GPL. The speechd-el package contains an Elisp library for communication with Speech Dispatcher, allowing you to write your own Emacs speech output extensions and your own Emacs interfaces to Speech Dispatcher. speechd-el design is completely different from the design of Emacspeak, another speech output package for Emacs. Though both speechd-el and Emacspeak serve a similar purpose, they differ in their approach to the problem they address. * What is new in the 0.5 version? - Reading of information about Emacs state (such as current buffer name and modes), including automatic notification of change states. - Mode and header lines contents can be read in CVS versions of Emacs. - It's possible to read messages displayed from built-in functions. - New user option speechd-speak-message-time-interval. - The SPEECHD_HOST environment variable is honored now. - New ways of communication with Speech Dispatcher. - The obsolete user option speechd-connection-parameters was removed. See the NEWS file for more details. * Where to get it? You can get the distribution tarball of the released version from http://www.freebsoft.org/pub/projects/speechd-el/speechd-el-0.5.tar.gz . The home page of the project is http://www.freebsoft.org/speechd-el . From pdm@freebsoft.org Tue Oct 19 19:43:58 2004 From: Milan Zamazal To: speechd-discuss@nongnu.org Date: Sun Aug 13 12:52:47 2006 Subject: speechd-el 0.5 released In-Reply-To: <87hdoqx1al.fsf@blackbird.zamazal.org> (Milan Zamazal's message of "Tue, 19 Oct 2004 15:52:50 +0200") References: <87hdoqx1al.fsf@blackbird.zamazal.org> Message-ID: <87zn2itxgh.fsf@blackbird.zamazal.org> The speechd-el-0.5.tar.gz file put on the web today was wrong. If you have downloaded it, please fetch it again, it's fixed now. With apology, Milan Zamazal From pdm@brailcom.org Fri Nov 12 13:23:24 2004 From: Milan Zamazal To: speechd-discuss@nongnu.org Date: Sun Aug 13 12:52:47 2006 Subject: Fw: Accessibility@freedesktop mailing list Message-ID: <87d5yjl0hv.fsf@blackbird.zamazal.org> An embedded message was scrubbed... From: Olaf Schmidt To: speechd-discuss@nongnu.org Subject: Accessibility@freedektop mailing list Date: Mon, 8 Nov 2004 11:21:33 +0100 Size: 5377 Url: http://lists.freebsoft.org/pipermail/speechd/attachments/20041112/93c52ed5/attachment.mht From seanm@dakotacom.net Sun Jan 9 01:49:46 2005 From: Sean McMahon To: speechd-discuss@nongnu.org Date: Sun Aug 13 12:52:47 2006 Subject: error encountered while trying to build speechd-up Message-ID: <20050109004946.GA2876@doghead> Using debian sarge with speakup cvs and a 2.4.27 custom kernel. I have speech-dispatcher already installed and have not yet created the softsynth device. speakup_sftsyn is a module in my kernel. see the attached file speechd-up_error.log This build of speechd-up came from cvs so I used the build.sh script to create the configure script. The error comes from the configure script. Thanks Sean -------------- next part -------------- checking for gcc... gcc checking for C compiler default output file name... a.out checking whether the C compiler works... yes checking whether we are cross compiling... no checking for suffix of executables... checking for suffix of object files... o checking whether we are using the GNU C compiler... yes checking whether gcc accepts -g... yes checking for gcc option to accept ANSI C... none needed checking for a BSD-compatible install... /usr/bin/install -c checking whether build environment is sane... yes checking whether make sets $(MAKE)... yes checking for working aclocal-1.4... found checking for working autoconf... found checking for working automake-1.4... found checking for working autoheader... found checking for working makeinfo... found checking for spd_open in -lspeechd... no *** Required speechd library missing! See INSTALL . From hanke@brailcom.org Sun Jan 9 12:23:18 2005 From: Hynek Hanke To: speechd-discuss@nongnu.org Date: Sun Aug 13 12:52:47 2006 Subject: error encountered while trying to build speechd-up In-Reply-To: <20050109004946.GA2876@doghead> References: <20050109004946.GA2876@doghead> Message-ID: <20050109112318.GD2578@brailcom.cz> On Sat, Jan 08, 2005 at 05:49:46PM -0700, Sean McMahon wrote: > Using debian sarge with speakup cvs and a 2.4.27 custom kernel. I have > speech-dispatcher already installed and have not yet created the > softsynth device. speakup_sftsyn is a module in my kernel. see the > attached file speechd-up_error.log > This build of speechd-up came from cvs so I used the build.sh script to > create the configure script. The error comes from the configure script. > [...] > *** Required speechd library missing! See INSTALL . You need to get the libspeechd1 and libspeechd-dev packages if you want to compile other programs with Speech Dispatcher support. With Regards, Hynek Hanke From seanm@dakotacom.net Mon Jan 10 09:16:26 2005 From: Sean McMahon To: speechd-discuss@nongnu.org Date: Sun Aug 13 12:52:47 2006 Subject: Another error while compiling speechd-up Message-ID: <20050110081626.GA7884@doghead> This time I got the speechdlib1 and speechd-dev packages. Configure ran fine for an error related to g++ which did not cause the script to exit. I install g++ ran make distclean, ran build.sh, ran ./configure and ran make all. This time the ./configure script did not give me any errors related to g++. Make all gave me the following error. See attached file speechd-upMake.log The error comes from function main and is the same error I received before installing g++ Once again this is coming from the cvs version I checkout yesterday. Sean -------------- next part -------------- gcc -DHAVE_CONFIG_H -I. -I. -I. -g -O2 -c speechd-up.c speechd-up.c: In function `main': speechd-up.c:421: error: `PIDPATH' undeclared (first use in this function) speechd-up.c:421: error: (Each undeclared identifier is reported only once speechd-up.c:421: error: for each function it appears in.) speechd-up.c:424: error: parse error before string constant make: *** [speechd-up.o] Error 1 From seanm@dakotacom.net Tue Jan 11 06:53:12 2005 From: Sean McMahon To: speechd-discuss@nongnu.org Date: Sun Aug 13 12:52:47 2006 Subject: more investigation into my compilation error with speechd-up Message-ID: <20050111055312.GA4260@doghead> This is a followup to my message from yesterday. I installed the flite-dev package and got the same error running make all. I noticed the configure checked for somethings I don't have specifically, a fortran compiler. Is this needed? I only saw .c files in the source code. Anyway here is a log of what happened during a run of the configure script. (see attached file configure.log) Thanks again for your help. If this has been fixed sinse saturday's checkout of the source code let me know. Sean -------------- next part -------------- checking for gcc... gcc checking for C compiler default output file name... a.out checking whether the C compiler works... yes checking whether we are cross compiling... no checking for suffix of executables... checking for suffix of object files... o checking whether we are using the GNU C compiler... yes checking whether gcc accepts -g... yes checking for gcc option to accept ANSI C... none needed checking for a BSD-compatible install... /usr/bin/install -c checking whether build environment is sane... yes checking whether make sets $(MAKE)... yes checking for working aclocal-1.4... found checking for working autoconf... found checking for working automake-1.4... found checking for working autoheader... found checking for working makeinfo... found checking for spd_open in -lspeechd... yes checking how to run the C preprocessor... gcc -E checking for egrep... grep -E checking for ANSI C header files... yes checking for sys/types.h... yes checking for sys/stat.h... yes checking for stdlib.h... yes checking for string.h... yes checking for memory.h... yes checking for strings.h... yes checking for inttypes.h... yes checking for stdint.h... yes checking for unistd.h... yes checking limits.h usability... yes checking limits.h presence... yes checking for limits.h... yes checking sys/ioctl.h usability... yes checking sys/ioctl.h presence... yes checking for sys/ioctl.h... yes checking sys/time.h usability... yes checking sys/time.h presence... yes checking for sys/time.h... yes checking for unistd.h... (cached) yes checking libspeechd.h usability... yes checking libspeechd.h presence... yes checking for libspeechd.h... yes checking for an ANSI C-conforming const... yes checking whether time.h and sys/time.h may both be included... yes checking whether gcc needs -traditional... no checking for select... yes checking for socket... yes checking for a BSD-compatible install... /usr/bin/install -c checking whether make sets $(MAKE)... (cached) yes checking build system type... i586-pc-linux-gnu checking host system type... i586-pc-linux-gnu checking for a sed that does not truncate output... /bin/sed checking for ld used by gcc... /usr/bin/ld checking if the linker (/usr/bin/ld) is GNU ld... yes checking for /usr/bin/ld option to reload object files... -r checking for BSD-compatible nm... /usr/bin/nm -B checking whether ln -s works... yes checking how to recognise dependent libraries... pass_all checking dlfcn.h usability... yes checking dlfcn.h presence... yes checking for dlfcn.h... yes checking for g++... g++ checking whether we are using the GNU C++ compiler... yes checking whether g++ accepts -g... yes checking how to run the C++ preprocessor... g++ -E checking for g77... no checking for f77... no checking for xlf... no checking for frt... no checking for pgf77... no checking for fort77... no checking for fl32... no checking for af77... no checking for f90... no checking for xlf90... no checking for pgf90... no checking for epcf90... no checking for f95... no checking for fort... no checking for xlf95... no checking for ifc... no checking for efc... no checking for pgf95... no checking for lf95... no checking for gfortran... no checking whether we are using the GNU Fortran 77 compiler... no checking whether accepts -g... no checking the maximum length of command line arguments... 32768 checking command to parse /usr/bin/nm -B output from gcc object... ok checking for objdir... .libs checking for ar... ar checking for ranlib... ranlib checking for strip... strip checking if gcc static flag works... yes checking if gcc supports -fno-rtti -fno-exceptions... no checking for gcc option to produce PIC... -fPIC checking if gcc PIC flag -fPIC works... yes checking if gcc supports -c -o file.o... yes checking whether the gcc linker (/usr/bin/ld) supports shared libraries... yes checking whether -lc should be explicitly linked in... no checking dynamic linker characteristics... GNU/Linux ld.so checking how to hardcode library paths into programs... immediate checking whether stripping libraries is possible... yes checking if libtool supports shared libraries... yes checking whether to build shared libraries... yes checking whether to build static libraries... yes configure: creating libtool appending configuration tag "CXX" to libtool checking for ld used by g++... /usr/bin/ld checking if the linker (/usr/bin/ld) is GNU ld... yes checking whether the g++ linker (/usr/bin/ld) supports shared libraries... yes checking for g++ option to produce PIC... -fPIC checking if g++ PIC flag -fPIC works... yes checking if g++ supports -c -o file.o... yes checking whether the g++ linker (/usr/bin/ld) supports shared libraries... yes checking dynamic linker characteristics... GNU/Linux ld.so checking how to hardcode library paths into programs... immediate checking whether stripping libraries is possible... yes appending configuration tag "F77" to libtool configure: creating ./config.status config.status: creating Makefile config.status: creating config.h config.status: executing default-1 commands From pdm@freebsoft.org Thu Jan 13 14:59:29 2005 From: Milan Zamazal To: speechd-discuss@nongnu.org Date: Sun Aug 13 12:52:47 2006 Subject: speechd-el 1.0 released Message-ID: <87u0pl77lq.fsf@blackbird.zamazal.org> speechd-el 1.0 released ======================= The Free(b)soft project is pleased to announce speechd-el 1.0. * What is speechd-el? speechd-el is an Emacs client to Speech Dispatcher, providing a complex speech interface to Emacs, focused especially on (but not limited to) the blind and visually impaired users. It allows the user to work with Emacs without looking on the screen, using the speech output produced by the synthesizers supported in Speech Dispatcher. Key speechd-el features are: - Automated speaking of most Emacs messages and actions, without necessity to customize each particular Emacs function to make it speak. - Most built-in and external Emacs packages can produce speech output immediately, without any special support. - Almost no change of the standard Emacs behavior, no interference with user settings. - Highly configurable. - Full cooperation with Speech Dispatcher. - Small code size. - Free Software, distributed under the GPL. The speechd-el package contains an Elisp library for communication with Speech Dispatcher, allowing you to write your own Emacs speech output extensions and your own Emacs interfaces to Speech Dispatcher. speechd-el design is completely different from the design of Emacspeak, another speech output package for Emacs. Though both speechd-el and Emacspeak serve a similar purpose, they differ in their approach to the problem they address. * What is new in the 1.0 version? There is no significant change in this release. It completes the development of the first stable version of speechd-el. After this stable release, major changes are planned in speechd-el, focused on providing more kinds of accessible user interfaces than just speech output. * Where to get it? You can get the distribution tarball of the released version from http://www.freebsoft.org/pub/projects/speechd-el/speechd-el-1.0.tar.gz . The home page of the project is http://www.freebsoft.org/speechd-el . From igueths@lava-net.com Thu Feb 10 03:58:06 2005 From: Igor Gueths To: speechd-discuss@nongnu.org Date: Sun Aug 13 12:52:47 2006 Subject: Speechd_up stopped working Message-ID: <20050210025806.GD25592@lava-net.com> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Hi all. I recently had to do an Fsck on my system since some Inodes decided to run away...Rather it was some bad blocks. They were fixed, or so I thought (the bad blocks all appeared within /usr). Ever since that point, Speechd_up says that it can't open /dev/softsynth. An ls -l /dev/softsynth gives me this: crw-r--r-- 1 root root 10, 26 2005-02-09 21:43 softsynth I can do an spd-say while Speech-dispatcher is running, and Flite works just fine. However, I get this in Speechd-up.log: [Wed Feb 9 21:46:10 2005] speechd: Speechd-speakup starts! [Wed Feb 9 21:46:10 2005] speechd: ERROR! Unable to open soft synth device (/dev/softsynth) Does anyone know why this might happen, considering that the softsynth device is properly created? I made sure to check that Sftsyn was compiled into my 2.6.6 kernel, which it appears to be. Does anyone have any ideas what might cause such strange behavior? Thanks! - -- The G-men all are cryin' And tearin' out their hair, 'Cause there's a new cryptography That's shown up everywhere. Nobody can break it, However good they be. Everybody's PC got the PGP. - ---The PGP song -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.2.6 (GNU/Linux) iD8DBQFCCs2+Nohoaf1zXJMRAmQSAJ9Gfi4pDsWsbK+IfPTREqwNjPkZrQCfZYUT v/JXQORJWs7zzoYc+ve0iSs= =JPNG -----END PGP SIGNATURE----- From Mikster4@msn.com Mon Mar 14 18:17:56 2005 From: Michael Whapples To: speechd-discuss@nongnu.org Date: Sun Aug 13 12:52:47 2006 Subject: speech-dispatcher stopped working Message-ID: Hello, I am using speech-dispatcher on a slackware installation. It compiled with no problems, and I was using it with speechd-up and speakup with no problems. That was until I plugged in an external sound card into my computer. It worked until I unplugged that USB sound card and rebooted. I am unable to start speech-dispatcher now. I am using speech-dispatcher 0.5. Can you help me get speech-dispatcher running again? From Michael Whapples "An optimist is someone who has never had much experience" From garycramblitt@comcast.net Fri Mar 25 19:10:28 2005 From: Gary Cramblitt To: speechd-discuss@nongnu.org Date: Sun Aug 13 12:52:47 2006 Subject: Find out speech has completed Message-ID: <200503251810.28744.garycramblitt@comcast.net> Is there any means to determine that a message has completed being spoken? I've looked over the API and don't see any events or callbacks. -- Gary Cramblitt (aka PhantomsDad) KDE Text-to-Speech Maintainer http://accessibility.kde.org/developer/kttsd/index.php From cerha@brailcom.org Tue Mar 29 16:18:49 2005 From: Tomas Cerha To: speechd-discuss@nongnu.org Date: Sun Aug 13 12:52:47 2006 Subject: [Fwd: Fw: Killed speech-dispatcher and can't get it running again] Message-ID: <424963C9.4060503@brailcom.org> -------- Original Message -------- Subject: Fw: Killed speech-dispatcher and can't get it running again Date: Fri, 18 Mar 2005 14:54:28 -0000 From: Michael Whapples To: speechd-discuss@nongnu.org To: Hello, I sent an e-mail to speechd@freebsoft.org about this, but I never got a response. Below is the e-mail I sent to the speakup mailing list regarding this. Any help would be useful. From Michael Whapples "An optimist is someone who has never had much experience" ----- Original Message ----- From: "Michael Whapples" To: speechd-discuss@nongnu.org To: "Speakup is a screen review system for Linux." Sent: Tuesday, March 15, 2005 6:14 PM Subject: Killed speech-dispatcher and can't get it running again > Hello, > I had speech-dispatcher running perfectly fine on my slackware > installation. That was until I connected my USB sound card. While the USB > sound card was connected, speech still came from the internal sound card. > After disconnecting the external sound card and rebooting, > speech-dispatcher does not want to run. It does not come up with any > errors, but it does not load. The internal sound card still works fine, I > have tested running flite from the command line, and also playing audio > files. I am using speech-dispatcher 0.5 with flite as output. Any help on > getting it running again would be welcome, and if anyone knows how to stop > it happening in the future, also let me know. > From > Michael Whapples > "An optimist is someone who has never had much experience" > _______________________________________________ > Speakup mailing list > Speakup@braille.uwo.ca > http://speech.braille.uwo.ca/mailman/listinfo/speakup > From buchal@brailcom.org Wed Mar 30 14:09:55 2005 From: Jan Buchal To: speechd-discuss@nongnu.org Date: Sun Aug 13 12:52:47 2006 Subject: New address for this list Message-ID: <86ekdx8hd8.fsf@brailcom.org> Hello, this list has a new email address: speechd@lists.freebsoft.org. Please send next emails on this address. Best regards -- Jan Buchal Tel: (00420) 224921679 Mob: (00420) 608023021 From pdm@zamazal.org Thu Mar 31 10:01:12 2005 From: Milan Zamazal To: speechd-discuss@nongnu.org Date: Sun Aug 13 12:52:48 2006 Subject: [Speechd] Re: Find out speech has completed Message-ID: <871x9wmegn.fsf@blackbird.zamazal.org> An embedded message was scrubbed... From: Milan Zamazal To: speechd-discuss@nongnu.org Subject: Re: Find out speech has completed Date: Tue, 29 Mar 2005 09:39:48 +0200 Size: 1825 Url: http://lists.freebsoft.org/pipermail/speechd/attachments/20050331/61fc2dfb/attachment.mht From root@mike Thu Mar 31 13:40:41 2005 From: root To: speechd-discuss@nongnu.org Date: Sun Aug 13 12:52:48 2006 Subject: [Speechd] Re: Killed speech-dispatcher and can't get it running again References: <424963C9.4060503@brailcom.org> <87vf78kzc2.fsf@blackbird.zamazal.org> Message-ID: Hello, I have managed to resolve this problem. The problem was that I had two versions of flite on my system, one from rpm's I found but they didn't seem to have the shared libaries, and one from a tgz package that some had built with speech-dispatcher in mind. After uninstalling the rpm version, and making sure the one from the tgz was installed it worked. For some reason it changed the version of flite it was working with, so it could not find shared libaries. From Michael Whapples Milan Zamazal writes: > > >> After disconnecting the external sound card and rebooting, > >> speech-dispatcher does not want to run. It does not come up with > >> any errors, but it does not load. The internal sound card still > >> works fine, I have tested running flite from the command line, > >> and also playing audio files. > > Hello Michael, > > I'm not an expert in this area, but I'd suggest to run Speech Dispatcher > as > > speech-dispatcher --run-single --log-level 5 > > and examine its log. Does Speech Dispatcher continue to run or does it > stop/crash? Anything interesting in the log? Anything interesting in > the debug log of the flite output module (turn it on in > /etc/speech-dispatcher/modules/flite.conf)? > > Regards, > > Milan Zamazal > > -- > I fear fear both faith and power > They have no need to use any question marks -- UJD > From verron@ircam.fr Mon Apr 18 19:46:13 2005 From: verron@ircam.fr To: speechd-discuss@nongnu.org Date: Sun Aug 13 12:52:47 2006 Subject: problem installing Message-ID: <43410.141.115.28.2.1113846373.squirrel@www.ircam.fr> Hi, I please need help. I am running a debian and I installed all the package mentionned in the INSTALL file. However it does not compile, as you can see here : charly@charlie:~/speechd-up$ ./configure checking for gcc... gcc checking for C compiler default output file name... a.out checking whether the C compiler works... yes checking whether we are cross compiling... no checking for suffix of executables... checking for suffix of object files... o checking whether we are using the GNU C compiler... yes checking whether gcc accepts -g... yes checking for gcc option to accept ANSI C... none needed checking for a BSD-compatible install... /usr/bin/install -c checking whether build environment is sane... yes checking whether make sets $(MAKE)... yes checking for working aclocal-1.4... found checking for working autoconf... found checking for working automake-1.4... found checking for working autoheader... found checking for working makeinfo... found checking for spd_open in -lspeechd... no *** Required speechd library missing! See INSTALL . The best, Charles Verron From kenny@hittsjunk.net Mon Apr 18 21:35:54 2005 From: Kenny Hitt To: speechd-discuss@nongnu.org Date: Sun Aug 13 12:52:47 2006 Subject: problem installing In-Reply-To: <43410.141.115.28.2.1113846373.squirrel@www.ircam.fr> References: <43410.141.115.28.2.1113846373.squirrel@www.ircam.fr> Message-ID: <20050418193554.GD28085@blackbox> Hi. Did you install the libspeechd-dev package? Hope this helps. Kenny From kristoffer.gustafsson@tranemo.org Wed Apr 27 14:38:04 2005 From: Gustafsson Kristoffer (elev) To: speechd-discuss@nongnu.org Date: Sun Aug 13 12:52:47 2006 Subject: help with speech dispatcher and festival Message-ID: <572BDE1B72FC114C8937BEA2691862B2165F07@titius.tranemo.se> Hello! I would like to get some help with festival and dispatcher. I've installed both and it works fine. I have tested speechd_up and it also works fine with festival. Now the problem is that I would like to use another voice than kal diphone voice. I've managed to change to the us english mbrola, but the voice I want is cepstral swift. That voice works with festival but I can't access from speech-dispatcher, or at least not with speechd_up. can you help me with this? /kristoffer -------------- next part -------------- An HTML attachment was scrubbed... URL: http://lists.freebsoft.org/pipermail/speechd/attachments/20050427/35445b41/attachment.html From pdm@freebsoft.org Thu Apr 28 15:35:16 2005 From: Milan Zamazal To: speechd-discuss@nongnu.org Date: Sun Aug 13 12:52:48 2006 Subject: help with speech dispatcher and festival In-Reply-To: <572BDE1B72FC114C8937BEA2691862B2165F07@titius.tranemo.se> (Gustafsson Kristoffer's message of "Wed, 27 Apr 2005 14:38:04 +0200") References: <572BDE1B72FC114C8937BEA2691862B2165F07@titius.tranemo.se> Message-ID: <8764y7yqgb.fsf@blackbird.zamazal.org> >>>>> "GK" == Gustafsson Kristoffer (elev) writes: GK> Now the problem is that I would like to use another voice than GK> kal diphone voice. I've managed to change to the us english GK> mbrola, but the voice I want is cepstral swift. That voice GK> works with festival but I can't access from speech-dispatcher, GK> or at least not with speechd_up. Could you please describe us how you configured the voice in Speech Dispatcher and what you exactly did to change the voice in Speech Dispatcher? Regards, Milan Zamazal -- As is typical in civilized (i.e., non-corporate) email discussions, my remarks *follow* (not precede) their context. -- Branden Robinson in debian-legal From garycramblitt@comcast.net Sun Apr 24 01:19:01 2005 From: Gary Cramblitt To: speechd-discuss@nongnu.org Date: Sun Aug 13 12:52:48 2006 Subject: [Speechd] KTTS and SpeechD integration Message-ID: <200504232319.01128.garycramblitt@comcast.net> Hynek and I have been discussing integration of the KDE Text-to-Speech System (KTTS) and Speech Dispatcher. If this could be done, it would offer several advantages: 1. KDE users would have TTS capability from boot-up to shutdown. (Currently, KDE must be running before KTTS can produce any speech.) 2. KDE users would have TTS capability in terminal (character cell) apps. 3. We could unify our efforts. If a new voice or synth became available, we need only enhance one package. 4. SpeechD performance and latency is better than KTTS. The eventual goal, if it can be achieved, is to eliminate the KTTS backend (called kttsd), and replace it with SpeechD. KTTS would then provide a GUI frontent for configuration of SpeechD, as well as additional capabilities, such as: 1. Ability for users to interactively pause, rewind, advance, or stop speech output. 2. Integration with the KDE notification system ("New mail has arrived."). 3. Text substitution and filtering. (The IRC message " Hello" becomes "PhantomsDad says Hello".) 4. Document conversion, such as HTML to SSML, PDF to SSML, etc. Towards this goal, I sat down to write a SpeechD plugin for KTTS, but immediately ran into some roadblocks. I'd like to explain these roadblocks so the SpeechD team can consider possible changes to SpeechD. KTTS uses a plugin architecture for synthesizers. The ideal plugin can asynchronously synthesize a message, notify KTTS when it is completed, and return a wav file. If a synth cannot return a wav file, the next ideal plugin asynchronously speaks a message, sending directly to audio device, and notifies KTTS when the speech output is finished. The next ideal plugin synchronously synthesizes a message and returns a wav file. The least ideal plugin synchronously synthesizes a message and sends it directly to the audio device. In order not to block KTTS or KDE apps, synchronous plugins are run in a separate thread. SpeechD doesn't fall into any of these models. It does not return a wav file. More seriously, it always runs asynchronously but does not notify when speech of a message has completed. KTTS parses text into individual sentences and sends them one at a time to the synth plugin. This is key in order to provide: 1. Ability to advance or rewind. 2. Ability to intermix with higher-priority messages. 3. Ability to change voice or synth in the middle of a long job. 4. Notification to apps of start/end of each sentence as well as text job as a whole. Now SpeechD has its own priority and queueing system, so my next approach was to forego these capabilities and immediately send all messages to SpeechD. In addition to losing the capabilities listed above, this would also mean that KTTS users could not combine SpeechD with other KTTS plugins, as speech from the other plugins would either block while SpeechD is speaking, or talk simultaneously, depending upon their PC's audio capabilities. KTTS provides 4 types/priorities of messages. In order of priority (highest to lowest) they are: Screen Reader. Interrupts all other messages, including other Screen Reader outputs. Not a queue; there is only one Screen Reader output at a time. Warning. Interrupts all lower-priority messages. Is a queue, so does not interrupt other Warnings. Message. Interrupts messages of type Text. Also a queue. Text. Interrupted by all other message types. A queue. Notice that none of these message types discard other messages except for Screen Reader, which only discards other Screen Reader messages. So I began looking at the message types/priorities in SpeechD API to see how the KTTS message types would map onto them. Since only Screen Reader discards other messages -- and only discards other Screen Reader messages, I immediately eliminated all the SpeechD message types that can be discarded -- namely 'Text', 'Notification', and 'Progress'. This left only 'Message' and 'Important'. So it appears I could map KTTS 'Text' messages to SpeechD 'Message' messages, and KTTS 'Message' and 'Warning' messages to SpeechD 'Important' messages. (We have considered eliminating 'Warning' type messages from KTTS anyway, so mapping both 'Warning' and 'Message' onto 'Important' would not be a hardship.) The following table summarizes: KTTS Type SpeechD Type ---------------- --------------------- Text Message Message Important Warning Important Screen Reader ? Now what to do about Screen Reader? The SpeechD message type that behaves most like KTTS Screen Reader is 'Text', but 'Text' messages are lower priority than 'Message' messages. Furthermore, 'Text' messages are discarded by 'Message' messages, but strangely, not discarded by 'Important' messages. Now it is possible I'm not reading the SpeechD API correctly. It may be that I am misinterpreting the word "cancel" in the docs. Under 'Important', it says -- When a new message of level `important' comes during a message of another priority is being spoken, this message other message is canceled and the message with priority `important' is said instead. Other messages of lower priorities are either postponed (priority `message' and `text') until there are no messages of priority important waiting or canceled (priority `notification' and `progress'. -- Then under 'Message' type it says -- If there are messages of priority `notification', `progress' or `text' waiting in the queue or being spoken when a message of priority `message' comes, these are canceled. -- Here, I interpret "canceled" as meaning discarded. Even if I have that wrong, and "canceled" just means postponed, it doesn't matter because 'Text' messages are of lower priority than 'Message' or 'Important' and therefore are not suitable for KTTS Screen Reader types. So what I need is a message type like 'Important', but which interrupts and discards itself. I thought about trying to use the SSIP CANCEL command to simulate such a message type, but since I have no way of knowing what kind of message SpeechD is currently speaking, that won't work. Stopping for a moment and reflecting on these issues, I came to the realization that SpeechD has a priority system that is ideal for Screen Readers, but not so good for speaking longer texts, such as web pages, pdf documents, or ebooks, while still providing interruption by higher-priority messages. The 'Text', 'Notification', and 'Progress' types are ideal for screen readers, but strangely are of lower priority than 'Important' or 'Message'. What seems to be missing is a "long text" type that is of lower priority than 'Text', 'Notification', and 'Progress', but is never discarded (unless application specifically cancels it.) Given these issues, I cannot presently move forward with integrating SpeechD with KTTS. If I had my way, SpeechD would offer callbacks to notify when messages have been spoken. This would allow me to immediately write a plugin for KTTS with the least amount of disruption to the existing KTTS architecture and API. It would also allow us to migrate the entire kttsd backend towards using SpeechD, although some additional changes to both APIs would be needed in order to accomplish that. Thanks for listening. -- Gary Cramblitt (aka PhantomsDad) KDE Text-to-Speech Maintainer http://accessibility.kde.org/developer/kttsd/index.php From pdm@freebsoft.org Thu Apr 28 18:30:25 2005 From: Milan Zamazal To: speechd-discuss@nongnu.org Date: Sun Aug 13 12:52:48 2006 Subject: [Speechd] KTTS and SpeechD integration In-Reply-To: <200504232319.01128.garycramblitt@comcast.net> (Gary Cramblitt's message of "Sat, 23 Apr 2005 23:19:01 +0000") References: <200504232319.01128.garycramblitt@comcast.net> Message-ID: <87mzriyice.fsf@blackbird.zamazal.org> Hello Gary, it's nice to see you work actively on the TTS front-ends all the time! >>>>> "GC" == Gary Cramblitt writes: GC> Hynek and I have been discussing integration of the KDE GC> Text-to-Speech System (KTTS) and Speech Dispatcher. If this GC> could be done, it would offer several advantages: It's an excellent idea! GC> Towards this goal, I sat down to write a SpeechD plugin for GC> KTTS, but immediately ran into some roadblocks. I'd like to GC> explain these roadblocks so the SpeechD team can consider GC> possible changes to SpeechD. Hynek is the one who should provide authoritative answers. But I can speak from a different point of view, the client side, as an author of a client extensively using Speech Dispatcher features (speechd-el for Emacs). GC> SpeechD doesn't fall into any of these models. It does not GC> return a wav file. I think this is because Speech Dispatcher is not intended to operate as a plugin to another message manager. I think its functionality is analogical to KTTS which perhaps neither returns a wave file. GC> More seriously, it always runs asynchronously but does not GC> notify when speech of a message has completed. This is a feature which should definitely be provided by Speech Dispatcher. We have been having it in mind for long time, but I don't know what the current state is. Hynek? GC> Now SpeechD has its own priority and queueing system, so my next GC> approach was to forego these capabilities and immediately send GC> all messages to SpeechD. In addition to losing the capabilities GC> listed above, this would also mean that KTTS users could not GC> combine SpeechD with other KTTS plugins, as speech from the GC> other plugins would either block while SpeechD is speaking, or GC> talk simultaneously, depending upon their PC's audio GC> capabilities. Again, I don't think using Speech Dispatcher as a plugin to another message manager is a good idea (although it can be useful for experiments right now). If the two systems were merged in future, some things might work better. GC> Now it is possible I'm not reading the SpeechD API correctly. GC> It may be that I am misinterpreting the word "cancel" in the GC> docs. Under 'Important', it says GC> -- When a new message of level `important' comes during a GC> message of another priority is being spoken, this message other GC> message is canceled and the message with priority `important' is GC> said instead. Other messages of lower priorities are either GC> postponed (priority `message' and `text') until there are no GC> messages of priority important waiting or canceled (priority GC> `notification' and `progress'. -- Yes, unless I'm very mistaken IMPORTANT just interrupts the messages and doesn't actually cancel them. The wording in the documentation is confusing. GC> Then under 'Message' type it says GC> -- If there are messages of priority `notification', `progress' GC> or `text' waiting in the queue or being spoken when a message of GC> priority `message' comes, these are canceled. -- This is right, in this case the messages are really cancelled. GC> So what I need is a message type like 'Important', but which GC> interrupts and discards itself. I thought about trying to use GC> the SSIP CANCEL command to simulate such a message type, but GC> since I have no way of knowing what kind of message SpeechD is GC> currently speaking, that won't work. You can do it by opening a separate Speech Dispatcher connection for the Screen Reader messages. Then CANCEL can apply only to messages sent through this connection, while the interruption still applies globally. But see below. GC> Stopping for a moment and reflecting on these issues, I came to GC> the realization that SpeechD has a priority system that is ideal GC> for Screen Readers, but not so good for speaking longer texts, GC> such as web pages, pdf documents, or ebooks, while still GC> providing interruption by higher-priority messages. The 'Text', GC> 'Notification', and 'Progress' types are ideal for screen GC> readers, but strangely are of lower priority than 'Important' or GC> 'Message'. What seems to be missing is a "long text" type that GC> is of lower priority than 'Text', 'Notification', and GC> 'Progress', but is never discarded (unless application GC> specifically cancels it.) Maybe it would be useful if we started with explanation of the background ideas of our message priority systems. If I understand the KTTS system right (please correct or complement me as necessary), it is based on the assumption that a user basically uses a screen reader for the purpose of navigating and controlling applications. This is a "single-task" highly interactive activity with immediate TTS feedback. Except this, system and applications can emit various more (Warning) or less (Message) urgent messages. Additionally, a user can start reading a longer text (the Text priority) on a background and continue doing other things. If I am correct above, I have some questions about the KTTS model: - When a user performs intensive screen reading, he can miss important messages like "the system is going down for reboot" or "phone call from Joe Hacker". The messages can be masked by the screen reading and the user can hear them only after it's too late. Do you rely on sound events? - How do you solve the problem of a growing queue when new messages are coming more quickly than they can be spoken? E.g. when a user types on the keyboard too quickly to be able to hear the character just typed in? - How do you ensure that a user doesn't hear too many uninteresting messages, i.e. messages which are only interesting now and not later and only in case nothing more important is to be spoken (e.g. reporting current time on background or a sequence of progress messages)? These questions provide motivation for the Speech Dispatcher priority system. The IMPORTANT priority serves exactly the purpose of signalling important system events (reboot, phone call, perhaps new mail) immediately, without breaking other messages. The other priorities are focused on common reading, they all work quite well together. Typically NOTIFICATION is used for low priority events which may be safely discarded if there is anything more interesting to say (e.g. typed characters, periodical clock or weather reporting). MESSAGE is for messages which should be spoken. TEXT is most common for regular reading. PROGRESS is good for sequences of progress messages, which you definitely don't want listen to all but which should be presented if nothing else to be spoken and after the progress report is finished. Long text reading can be done simply with the TEXT priority assuming the user listens to the text and doesn't do anything else. I think "background" text reading is not possible in the Speech Dispatcher model. I like your "long text" priority idea, it might be useful to add it in some form to Speech Dispatcher. GC> Thanks for listening. Thanks for your observations and suggestions. Regards, Milan Zamazal -- The world is not something you can wrap your head around without needing years of experience. -- Kent M. Pitman in comp.lang.lisp From garycramblitt@comcast.net Thu Apr 28 16:30:01 2005 From: Gary Cramblitt To: speechd-discuss@nongnu.org Date: Sun Aug 13 12:52:48 2006 Subject: [Speechd] KTTS and SpeechD integration In-Reply-To: <87mzriyice.fsf@blackbird.zamazal.org> References: <200504232319.01128.garycramblitt@comcast.net> <87mzriyice.fsf@blackbird.zamazal.org> Message-ID: <200504281430.01146.garycramblitt@comcast.net> On Thursday 28 April 2005 04:30 pm, Milan Zamazal wrote: > Hello Gary, > > it's nice to see you work actively on the TTS front-ends all the time! > > >>>>> "GC" == Gary Cramblitt writes: > > GC> Hynek and I have been discussing integration of the KDE > GC> Text-to-Speech System (KTTS) and Speech Dispatcher. If this > GC> could be done, it would offer several advantages: > > It's an excellent idea! > > GC> Towards this goal, I sat down to write a SpeechD plugin for > GC> KTTS, but immediately ran into some roadblocks. I'd like to > GC> explain these roadblocks so the SpeechD team can consider > GC> possible changes to SpeechD. > > Hynek is the one who should provide authoritative answers. But I can > speak from a different point of view, the client side, as an author of a > client extensively using Speech Dispatcher features (speechd-el for > Emacs). > > GC> SpeechD doesn't fall into any of these models. It does not > GC> return a wav file. > > I think this is because Speech Dispatcher is not intended to operate as > a plugin to another message manager. I understand. Also, returning a wav file adds latency. This is a tricky problem for which I have no good answer today. KDE is going to be changing its multi-media architecture and integrating that with Speech Dispatcher will present some challenges. For example, I can't tell you whether SpeechD should use GStreamer, or ALSA, or ... whatever. I don't think returning a wav file is essential at this time, but it might become essential in the future. For now, I'm treating it as an issue that can't be solved today, but must be resolved at some point. > I think its functionality is > analogical to KTTS which perhaps neither returns a wave file. True, although KTTS could be easily enhanced to do that. > > GC> More seriously, it always runs asynchronously but does not > GC> notify when speech of a message has completed. > > This is a feature which should definitely be provided by Speech > Dispatcher. We have been having it in mind for long time, but I don't > know what the current state is. Hynek? > > GC> Now SpeechD has its own priority and queueing system, so my next > GC> approach was to forego these capabilities and immediately send > GC> all messages to SpeechD. In addition to losing the capabilities > GC> listed above, this would also mean that KTTS users could not > GC> combine SpeechD with other KTTS plugins, as speech from the > GC> other plugins would either block while SpeechD is speaking, or > GC> talk simultaneously, depending upon their PC's audio > GC> capabilities. > > Again, I don't think using Speech Dispatcher as a plugin to another > message manager is a good idea (although it can be useful for > experiments right now). If the two systems were merged in future, some > things might work better. Agreed. Yes, the plugin is a step towards the longer range goal of eliminating the KTTSD backend entirely, leaving a DCOP wrapper, KDE notification interface, enhanced filtering and document conversion, and a GUI for configuring everything and controlling speech in real time (Pause, Resume, Cancel). The KTTSD queueing and prioritization would be eliminated. KTTS would no longer be a message manager, but would add additional capabilities. It is even conceivable that KDE programmers might want to bypass the DCOP interface and interface directly with SpeechD. Time will tell. To the extent that writing a SpeechD plugin for KTTS raises issues, such as those we are discussing here, it is a good thing. In a private email to Hynek I stated that the SpeechD plugin for KTTS would be distributed as "experimental" and not a final or even recommended solution. > > GC> Now it is possible I'm not reading the SpeechD API correctly. > GC> It may be that I am misinterpreting the word "cancel" in the > GC> docs. Under 'Important', it says > > GC> -- When a new message of level `important' comes during a > GC> message of another priority is being spoken, this message other > GC> message is canceled and the message with priority `important' is > GC> said instead. Other messages of lower priorities are either > GC> postponed (priority `message' and `text') until there are no > GC> messages of priority important waiting or canceled (priority > GC> `notification' and `progress'. -- > > Yes, unless I'm very mistaken IMPORTANT just interrupts the messages and > doesn't actually cancel them. The wording in the documentation is > confusing. > > GC> Then under 'Message' type it says > > GC> -- If there are messages of priority `notification', `progress' > GC> or `text' waiting in the queue or being spoken when a message of > GC> priority `message' comes, these are canceled. -- > > This is right, in this case the messages are really cancelled. > > GC> So what I need is a message type like 'Important', but which > GC> interrupts and discards itself. I thought about trying to use > GC> the SSIP CANCEL command to simulate such a message type, but > GC> since I have no way of knowing what kind of message SpeechD is > GC> currently speaking, that won't work. > > You can do it by opening a separate Speech Dispatcher connection for the > Screen Reader messages. Then CANCEL can apply only to messages sent > through this connection, while the interruption still applies globally. > But see below. If I did SCREEN READER in a separate connection, would the two connections speak simultaneously (not necessarily a bad thing)? > > GC> Stopping for a moment and reflecting on these issues, I came to > GC> the realization that SpeechD has a priority system that is ideal > GC> for Screen Readers, but not so good for speaking longer texts, > GC> such as web pages, pdf documents, or ebooks, while still > GC> providing interruption by higher-priority messages. The 'Text', > GC> 'Notification', and 'Progress' types are ideal for screen > GC> readers, but strangely are of lower priority than 'Important' or > GC> 'Message'. What seems to be missing is a "long text" type that > GC> is of lower priority than 'Text', 'Notification', and > GC> 'Progress', but is never discarded (unless application > GC> specifically cancels it.) > > Maybe it would be useful if we started with explanation of the > background ideas of our message priority systems. > > If I understand the KTTS system right (please correct or complement me > as necessary), it is based on the assumption that a user basically uses > a screen reader for the purpose of navigating and controlling > applications. This is a "single-task" highly interactive activity with > immediate TTS feedback. Except this, system and applications can emit > various more (Warning) or less (Message) urgent messages. Additionally, > a user can start reading a longer text (the Text priority) on a > background and continue doing other things. Yes, that's good summary of the existing KTTS model. > > If I am correct above, I have some questions about the KTTS model: > > - When a user performs intensive screen reading, he can miss important > messages like "the system is going down for reboot" or "phone call > from Joe Hacker". The messages can be masked by the screen reading > and the user can hear them only after it's too late. Yes, that would be a problem. > Do you rely on sound events? There is a capability to use sound when long text is interrupted by SCREEN READER, WARNING, or MESSAGE. User must configure this however. Sound events are on the TODO list. > > - How do you solve the problem of a growing queue when new messages are > coming more quickly than they can be spoken? E.g. when a user types > on the keyboard too quickly to be able to hear the character just > typed in? I guess we don't. If the keyboard app uses SCREEN READER, then fast typing will discard speaking some of the letters. If the keyboard app uses WARNING or MESSAGE, they will not be discarded. We will probably want to add a NOTIFICATION type to address this. > > - How do you ensure that a user doesn't hear too many uninteresting > messages, i.e. messages which are only interesting now and not later > and only in case nothing more important is to be spoken > (e.g. reporting current time on background or a sequence of progress > messages)? We will probably want to add a message type similar to PROGRESS. > > These questions provide motivation for the Speech Dispatcher priority > system. The IMPORTANT priority serves exactly the purpose of signalling > important system events (reboot, phone call, perhaps new mail) > immediately, without breaking other messages. The other priorities are > focused on common reading, they all work quite well together. Typically > NOTIFICATION is used for low priority events which may be safely > discarded if there is anything more interesting to say (e.g. typed > characters, periodical clock or weather reporting). MESSAGE is for > messages which should be spoken. TEXT is most common for regular > reading. PROGRESS is good for sequences of progress messages, which you > definitely don't want listen to all but which should be presented if > nothing else to be spoken and after the progress report is finished. As KDE does not currently have a Screen Reader, we haven't had to address the issues it raises as well as Speech Dispatcher has. I anticipate that we will want to incorporate much of the SpeechD model. In fact, in the longer range solution, it will *be* the SpeechD model since SpeechD will do all the queueing and prioritization. > > Long text reading can be done simply with the TEXT priority assuming the > user listens to the text and doesn't do anything else. I think > "background" text reading is not possible in the Speech Dispatcher > model. I like your "long text" priority idea, it might be useful to add > it in some form to Speech Dispatcher. Yes please. -- Gary Cramblitt (aka PhantomsDad) KDE Text-to-Speech Maintainer http://accessibility.kde.org/developer/kttsd/index.php From pdm@freebsoft.org Fri Apr 29 14:14:32 2005 From: Milan Zamazal To: speechd-discuss@nongnu.org Date: Sun Aug 13 12:52:48 2006 Subject: [Speechd] KTTS and SpeechD integration In-Reply-To: <200504281430.01146.garycramblitt@comcast.net> (Gary Cramblitt's message of "Thu, 28 Apr 2005 14:30:01 +0000") References: <200504232319.01128.garycramblitt@comcast.net> <87mzriyice.fsf@blackbird.zamazal.org> <200504281430.01146.garycramblitt@comcast.net> Message-ID: <87r7gtg4pj.fsf@blackbird.zamazal.org> >>>>> "GC" == Gary Cramblitt writes: GC> KDE is going to be changing its multi-media architecture and GC> integrating that with Speech Dispatcher will present some GC> challenges. For example, I can't tell you whether SpeechD GC> should use GStreamer, or ALSA, or ... whatever. Generally, I wouldn't worry much about particular implementation features, they can be added any time. The difficult thing is design. Maybe you can work with Hynek on adding special unofficial Speech Dispatcher features necessary for the step-by-step KDE integration. GC> If I did SCREEN READER in a separate connection, would the two GC> connections speak simultaneously (not necessarily a bad thing)? No. It's expected that each application connects to Speech Dispatcher separately, while the message queue is global. >> Long text reading can be done simply with the TEXT priority >> assuming the user listens to the text and doesn't do anything >> else. I think "background" text reading is not possible in the >> Speech Dispatcher model. I like your "long text" priority idea, >> it might be useful to add it in some form to Speech Dispatcher. GC> Yes please. What do you think, Hynek? Regards, Milan Zamazal -- Here is my advice, don't try to program the bleeding edge for the general populace unless you really, really, really like migraines. Neal H. Walfield From hanke@brailcom.org Wed May 4 23:18:37 2005 From: Hynek Hanke To: speechd-discuss@nongnu.org Date: Sun Aug 13 12:52:48 2006 Subject: [Speechd] KTTS and SpeechD integration In-Reply-To: <200504232319.01128.garycramblitt@comcast.net> References: <200504232319.01128.garycramblitt@comcast.net> Message-ID: <20050504211837.GA2246@brailcom.cz> Hello Gary, thank you for your detailed description of the goals and remaining issues. I'll try to comment on the later. 1) Synchronization with client applications > If a synth cannot return a wav file, the next ideal > plugin asynchronously speaks a message, sending directly to audio device, and > notifies KTTS when the speech output is finished. I think this is the way we can go. Speech Dispatcher currently doesn't support notification, but many things are already prepared so that this can be implemented. Speech dispatcher has all the necessary information internally, so the only remaining issue is how to communicate this information to the client application. When that is solved/implemented, it will notify client applications not only about the beginning and the end of speaking a message, but also about reaching index marks, which you can insert into SSML messages (when the end synthesizer allows). I'll work on it now. I'm going to post a separate email about this soon. 2) Sentence boundary detection > KTTS parses text into individual sentences and sends them one at a time to the > synth plugin. This is key in order to provide: We have also been doing this in Speech Dispatcher some time ago, but it turned out not to be the best approach for several reasons I'll explain bellow. I believe festival-freebsoft-utils now offers a better solutions which, in connection with index marking, could address these goals too. There is a number of issues we found with cutting text into sentences in Speech Dispatcher and sending just sentences to the output modules and synthesizers. 1) The TTS has no possibility to do more advanced syntactic analysis. It is only allowed to operate on one sentence. 2) We need to handle language dependent issues in a project (Speech Dispatcher, KTTSD) that should be language independent. 3) How to cut SSML or other markup into sentences? 4) How to cut data that are not an ordinary text (program source code, ...) 5) It makes the output module much more complicated if good performance is of concern. It's necessary to already have sent for synthesis the next sentence before the previous one is spoken in the speakers so that the TTS doesn't sit idle. Sentences of different length may cause unnecessary delays. We found it would be much better, and it's also natural, to pass the whole text into the TTS and let the TTS do this job instead. The TTS needs to somehow cut the text into smaller chunks anyway, because of index marks and because of performance reasons, but it can do it only after the necessary SSML parsing and syntax analysis has been done, it can do it in a language dependent manner and it can do it according to it's own taste (which allows TTS programmers to implement algorithms for better performance etc.) It should not be a problem to pass the whole text into the TTS at once as long as: 1) The TTS is able to return partial results as soon as they are available (so that we don't have to wait untill the whole text is synthesized). 2) The TTS provides synchronization information (index marks). 3) The TTS is able to start synthesizing the text from an arbitrary index mark in the given (complete) text. Currently, (1) and (2) are provided by festival-freebsoft-utils. What do you Milan think about (3)? Let me explain how these three would address your needs. > 1. Ability to advance or rewind. The output module has the information about the position in the text from receiving the index marks (2). It can skip sentences forwards or backwards by sending the whole text again together with the identification number of index mark you it wants to start from, according to (3). > 2. Ability to intermix with higher-priority messages. When a higher-priority message comes, the output module knows the position from the last received index mark (2), so it can instruct the TTS to start playing there again (3) when that higher-priority message is spoken. > 3. Ability to change voice or synth in the middle of a long job. This is very similar to the previous point. > 4. Notification to apps of start/end of each sentence as well as text job > as a whole. Yes, this is very desirable and (2) ensures it's possible. Currently, this is implemented between Speech Dispatcher and Festival in the following way. Speech Dispatcher still does some basic sentence boundary detection, but only to insert some kind of private index marks (additionally to index marks inserted by the client application). Ideally, this should also somehow be handled by Festival or the other TTS in use in the future so that Speech Dispatcher doesn't have to do anything with the text itself. Then the whole text (in SSML) is sent to Festival for synthesis and the function (speechd-next) Festival function is repeatedly called to retrieve the newly synthesized text or the information about an index mark being reached. 3) Priority models > KTTS Type SpeechD Type > ---------------- --------------------- > Text Message > Message Important > Warning Important This mapping would go strongly against the philosophy of Speech Dispatcher, because the priority important is really only to be used for short important messages, not for ordinary messages, to prevent ``polution'' of the queues with important messages which can't be discarded nor postponed by any other priority. Let me stop for a moment and explain our view on the message management in accessibility. We don't imagine accessible computer for the blind as just a screen reader, we also think about the concept of what we currently call application reader. I'll explain the difference. A screen reader is a general purpose tool which you can use to do most of your work and it should provide you a reasonable level of access to all aplications that don't have some kind of barier in themselves. Screen reader typically only sees the surface of what actually is contained in the application. It might be customized to some degree for that particular application, but still it doesn't have access to some information the application knows. So it might make sense to build accessibility to some applications directly, so that these accessibility solutions can take advantage of the particular details of the application and of the information that are available to it when it operates *inside* the application. I'll give two examples. One of them is what we are currently doing with Emacs in speechd-el. speechd-el knows the origin of each message so it can decide which information is and which isn't important for the user (according to user configuration of course) and take advantage of the advanced priority model. It knows the language and character encoding of buffers, so it can switch languages automatically. It can perform other actions that a general purpose screen reader couldn't. Maybe Milan Zamazal can give better examples. Another example is the GNU Typist package. If you are not familiar with, it's a very good tool to help people learn touch typing. While in the previous case of Emacs, building accessibility into the application directly and not relying on a general purpose screen reader was a convenience, for GNU Typist that will be an absolute need. I can't imagine how would you be trying to learn touch typing just by reading the application with a general purpose screen reader -- you need dictation, you need notification about errors and so on and so on, something that a general purpose screen reader without any understanding of the application can't provide. Of course these specialized solutions, which we call application readers, should not be hardwired into the main program. Rather, they should be some extension or an additional layer built on top of the program. Let's return to the priority model. Now you understand that once we bypass the earliest stage in the GUI, we will get to the situation that we already are in textmode. There will typically be more clients connected to the high level TTS API in use (KTTSD, Speech Dispatcher), not just the screen reader. Some of them might be able to make a good use of the whole priority system, some not. Now you come to understand why we don't strictly separate the screen reader and others in Speech Dispatcher and why we don't have any priority that an application could use to have a full and complete controll over everything said. Rather, applications are supposed to use priorities as text and message and when such a situation happens that there is some more important message to be spoken (either originating in these applications or somewhere else), than this message *should* pass through and the original messages will be postponed or discarded. I think thinking in these terms should be the basis for our future work. Obviously, the priority system in Speech Dispatcher is not optimal and we will have to continue working on it. Milan explained the ideas under the different priorities in a separate email, I tried to explain the motivation for them. I very much welcome all your comments about all of this. I really like the ``long text'' idea. Maybe the nicest behavior would be if every incomming higher-priority message would just pause reading ``long text'' and then resume it again after the higher-priority message is spoken. This way, a user could be reading an ebook while still being able to listen to time and new-email notifications, being able to jump into the mixer to adjust the sound volume etc. What do you think of it? 4) Conclusions > Given these issues, I cannot presently move forward with integrating SpeechD > with KTTS. If I had my way, SpeechD would offer callbacks to notify when > messages have been spoken. It will. What to do with KTTSD ``Screen Reader'' priority is still an open issue. Thank you & please send your comments, Hynek From pdm@freebsoft.org Thu May 5 00:38:28 2005 From: Milan Zamazal To: speechd-discuss@nongnu.org Date: Sun Aug 13 12:52:48 2006 Subject: [Speechd] KTTS and SpeechD integration In-Reply-To: <20050504211837.GA2246@brailcom.cz> (Hynek Hanke's message of "Wed, 4 May 2005 23:18:37 +0200") References: <200504232319.01128.garycramblitt@comcast.net> <20050504211837.GA2246@brailcom.cz> Message-ID: <87d5s6borf.fsf@blackbird.zamazal.org> >>>>> "HH" == Hynek Hanke writes: HH> 3) The TTS is able to start synthesizing the text from an HH> arbitrary index mark in the given (complete) text. HH> What do you Milan think about (3)? I'm not sure what's the point here. I assume you don't think about just cutting out the before-mark prefix, you probably think about starting the synthesis considering the left context, e.g. correct pronunciation ("I will read" vs. "I have read"), intonation dependent on the previous contents, or considering surrounding SSML markup. I think this is a legitimate requirement on a TTS system. But retaining the complete left context means processing it, which may mean the synthesizer doesn't start returning the output "immediately". Solving the related problems may not be easy. Anyway, yes, it should be a job of the TTS system, which can understand the text to the extent of the needs of its speech synthesis mechanism. HH> A screen reader is a general purpose tool which you can use to HH> do most of your work and it should provide you a reasonable HH> level of access to all aplications that don't have some kind of HH> barier in themselves. HH> Screen reader typically only sees the surface of what actually HH> is contained in the application. It might be customized to some HH> degree for that particular application, but still it doesn't HH> have access to some information the application knows. HH> So it might make sense to build accessibility to some HH> applications directly, so that these accessibility solutions can HH> take advantage of the particular details of the application and HH> of the information that are available to it when it operates HH> *inside* the application. HH> I'll give two examples. HH> speechd-el knows the origin of each message so it can decide HH> which information is and which isn't important for the user HH> (according to user configuration of course) and take advantage HH> of the advanced priority model. It knows the language and HH> character encoding of buffers, so it can switch languages HH> automatically. It can perform other actions that a general HH> purpose screen reader couldn't. What you say here is probably somewhat confused. speechd-el is a sort of screen reader inside the Emacs desktop. Generally it knows nothing about origins of messages nor about particular applications implemented in Emacs. It just utilizes information Emacs provides about the text to make a (mostly) proper alternative output. This is a very important design decision, which makes speechd-el working better and without maintenance nightmares, unlike highly application specific accessibility solutions. No Emacs application is modified for the purpose of smooth working with speechd-el (not counting user customizations). In my point of view the difference between a "screen reader" and an "application reader" is that a screen reader reads contents of the screen, while an application reader reads data provided by the application. From this point of view both speechd-el and what Gary was talking about are application readers; but from the point of view of your definition both are screen readers. As for language and coding information: speechd-el doesn't know the language of the text, unless it receives some hint; this is not much different from what any common screen(Hynek)/application(Milan) reader could do. speechd-el doesn't care about coding at all, it works on Emacs *characters*; speechd-el just asks Emacs to send the characters to the alternative output device (Speech Dispatcher, BrlTTY) in the given coding. Please note speechd-el is not the final accessibility solution for the _application_ called Emacs. It is a practical demonstration of a general accessibility solution for the Emacs _desktop_. Also (more importantly) it is an intermediate *working* solution providing wide range of accessible applications to a user until a more general good, powerful and stable application accessibility interface is defined and implemented. Once it becomes available, speechd-el can turn into the interface bindings (of the Emacs _application_) and won't communicate with Speech Dispatcher itself any longer. The Emacs independent application reader will do. I think making applications accessible doesn't have much to do with message priorities. OTOH the Speech Dispatcher priority model proved to be very helpful in speechd-el, for both the speech and Braille outputs. So I think the priority model is suitable for that kind of screen/application readers. Regards, Milan Zamazal -- Free software is about freedom, not about free beer. If you care only about the latter, you'll end up with no freedom and no free beer. From hanke@brailcom.org Fri May 6 21:16:14 2005 From: Hynek Hanke To: speechd-discuss@nongnu.org Date: Sun Aug 13 12:52:48 2006 Subject: [Speechd] KTTS and SpeechD integration In-Reply-To: <87d5s6borf.fsf@blackbird.zamazal.org> References: <200504232319.01128.garycramblitt@comcast.net> <20050504211837.GA2246@brailcom.cz> <87d5s6borf.fsf@blackbird.zamazal.org> Message-ID: <20050506191243.GB2959@brailcom.cz> Hi Milan, > HH> 3) The TTS is able to start synthesizing the text from an > HH> arbitrary index mark in the given (complete) text. > > I'm not sure what's the point here. I assume you don't think about just > cutting out the before-mark prefix, you probably think about starting > the synthesis considering the left context, e.g. correct pronunciation > ("I will read" vs. "I have read"), intonation dependent on > the previous contents, or considering surrounding SSML markup. you understood it correctly. I don't think considering pronounciation and intonation context is that important, especially because (except for special cases), this kind of index marks will mostly lie in between sentences, not inside them. But considering surrounding SSML markup is very important, because that can make a big difference in the resulting speech. Currently, what I do in Speech Dispatcher to ensure pause/resume or rewinding, is that I cut the message at the appropriate index mark and send to Festival only the rest. This is very wrong, because it can even break the SSML entirely (make it invalid) if the index mark happens to be at some bad place. Fixing this is of quite high importance, I think. I didn't think about this problem when we made the switch to SSML. > I think this is a legitimate requirement on a TTS system. I think it should also be somehow included in the TTS API specs discussed on freedesktop.org. The point 4.7 (NICE TO HAVE) is somehow similar to this requirement, although a bit more general. The problem with this is that if the synthesizer supports SSML and doesn't support the feature described above, then the applications are forced to parse the SSML if they want to do pause/resume without blocking the whole session with the TTS (that's what the pause/resume functions in 4.12 do, if I understand correctly). I propose to prepend a SHOULD HAVE point requiring the TTS to at least be able to start reading from a specified index mark, as described above. > But retaining the complete left context means processing it, which may mean > the synthesizer doesn't start returning the output "immediately". If this turns out to be a problem, we could look for alternative ways how to ensure pause/resume/rewinding, like the point 4.7 of TTS API requirements (although this point is not complete for that purpose). But I expect it would be more difficult. > What you say here is probably somewhat confused. Of course your explanations of the screen/application reader are more acurate. My point was just to explain that generally a reader like Gnopernicus won't be the only application that needs it's speech to get through so that accessibility is ensured. There even might be more important messages that what gets out of Gnopernicus or speechd-el. A user probably doesn't want the message ``You have an incomming call from Milan Zamazal'' be interrupted or thrown away by messages caused by him writing a letter in a text editor. So I don't think screen readers or application readers should be, by default, in power over everything. With Regards, Hynek Hanke From pdm@freebsoft.org Sun May 8 17:52:59 2005 From: Milan Zamazal To: speechd-discuss@nongnu.org Date: Sun Aug 13 12:52:48 2006 Subject: [Speechd] KTTS and SpeechD integration In-Reply-To: <20050506191243.GB2959@brailcom.cz> (Hynek Hanke's message of "Fri, 6 May 2005 21:16:14 +0200") References: <200504232319.01128.garycramblitt@comcast.net> <20050504211837.GA2246@brailcom.cz> <87d5s6borf.fsf@blackbird.zamazal.org> <20050506191243.GB2959@brailcom.cz> Message-ID: <874qdd90kk.fsf@blackbird.zamazal.org> >>>>> "HH" == Hynek Hanke writes: HH> I think it should also be somehow included in the TTS API specs HH> discussed on freedesktop.org. The point 4.7 (NICE TO HAVE) is HH> somehow similar to this requirement, although a bit more HH> general. [...] HH> I propose to prepend a SHOULD HAVE point requiring the TTS to at HH> least be able to start reading from a specified index mark, as HH> described above. Please send the suggestion together with a proper explanation of the motivation to the appropriate freedesktop.org mailing list. HH> My point was just to explain that generally a reader like HH> Gnopernicus won't be the only application that needs it's speech HH> to get through so that accessibility is ensured. There even HH> might be more important messages that what gets out of HH> Gnopernicus or speechd-el. HH> A user probably doesn't want the message ``You have an incomming HH> call from Milan Zamazal'' be interrupted or thrown away by HH> messages caused by him writing a letter in a text editor. IMHO applications shouldn't decide about importance or the particular way of presentation of such messages. They should just provide enough information for such decisions to application readers (such as Gnopernicus). Regards, Milan Zamazal -- I fear fear both faith and power They have no need to use any question marks -- UJD From garycramblitt@comcast.net Mon May 9 05:06:12 2005 From: Gary Cramblitt To: speechd-discuss@nongnu.org Date: Sun Aug 13 12:52:48 2006 Subject: [Speechd] KTTS and SpeechD integration In-Reply-To: <20050504211837.GA2246@brailcom.cz> References: <200504232319.01128.garycramblitt@comcast.net> <20050504211837.GA2246@brailcom.cz> Message-ID: <200505082306.12084.garycramblitt@comcast.net> Sorry for the late reply. I've been busy with some other things. I'm very encouraged by your reply and believe we can reach a mutually acceptable solution. On Wednesday 04 May 2005 05:18 pm, Hynek Hanke wrote: > Hello Gary, > > thank you for your detailed description of the goals and remaining issues. > I'll try to comment on the later. > > 1) Synchronization with client applications > > > If a synth cannot return a wav file, the next ideal > > plugin asynchronously speaks a message, sending directly to audio device, > > and notifies KTTS when the speech output is finished. > > I think this is the way we can go. Speech Dispatcher currently doesn't > support notification, but many things are already prepared so that this can > be implemented. Speech dispatcher has all the necessary information > internally, so the only remaining issue is how to communicate this > information to the client application. When that is solved/implemented, it > will notify client applications not only about the beginning and the end of > speaking a message, but also about reaching index marks, which you can > insert into SSML messages (when the end synthesizer allows). > > I'll work on it now. I'm going to post a separate email about this soon. I'm willing to forego the requirement to return a wav file if SpeechD can offer the functionality needed. I have some suggestions for how we might want to implement callbacks. I'll send in a separate email to this list. > > 2) Sentence boundary detection > > > KTTS parses text into individual sentences and sends them one at a time > > to the synth plugin. This is key in order to provide: > > We have also been doing this in Speech Dispatcher some time ago, but it > turned out not to be the best approach for several reasons I'll explain > bellow. I believe festival-freebsoft-utils now offers a better solutions > which, in connection with index marking, could address these goals too. > > There is a number of issues we found with cutting text into sentences in > Speech Dispatcher and sending just sentences to the output modules and > synthesizers. > > 1) The TTS has no possibility to do more advanced syntactic analysis. It is > only allowed to operate on one sentence. > > 2) We need to handle language dependent issues in a project (Speech > Dispatcher, KTTSD) that should be language independent. > > 3) How to cut SSML or other markup into sentences? > > 4) How to cut data that are not an ordinary text (program source code, ...) > > 5) It makes the output module much more complicated if good performance is > of concern. It's necessary to already have sent for synthesis the next > sentence before the previous one is spoken in the speakers so that the TTS > doesn't sit idle. Sentences of different length may cause unnecessary > delays. KTTS actually addresses all these issues fairly well. I'll explain in a separate email, but it is academic because KTTS can forego Sentence Boundary Detection so long as needed functionality is provided by SpeechD. > > It should not be a problem to pass the whole text into the TTS at once as > long as: > > 1) The TTS is able to return partial results as soon as they are available > (so that we don't have to wait untill the whole text is synthesized). > > 2) The TTS provides synchronization information (index marks). > > 3) The TTS is able to start synthesizing the text from an arbitrary index > mark in the given (complete) text. > > Currently, (1) and (2) are provided by festival-freebsoft-utils. What do > you Milan think about (3)? > > Let me explain how these three would address your needs. > > > 1. Ability to advance or rewind. > > The output module has the information about the position in the text from > receiving the index marks (2). It can skip sentences forwards or backwards > by sending the whole text again together with the identification number of > index mark you it wants to start from, according to (3). In the case of advance/rewind of text that is not yet completely spoken, you might want to consider providing a way to do this without the need to resend text, since I'm assuming that SpeechD still has the text. Otherwise, KTTS would need to maintain a queue of its own. My objective is to remove all queuing from KTTS and let SpeechD handle that. > > > 2. Ability to intermix with higher-priority messages. > > When a higher-priority message comes, the output module knows the position > from the last received index mark (2), so it can instruct the TTS to start > playing there again (3) when that higher-priority message is spoken. > > > 3. Ability to change voice or synth in the middle of a long job. > > This is very similar to the previous point. > > > 4. Notification to apps of start/end of each sentence as well as text > > job as a whole. > > Yes, this is very desirable and (2) ensures it's possible. > > Currently, this is implemented between Speech Dispatcher and Festival in > the following way. Speech Dispatcher still does some basic sentence > boundary detection, but only to insert some kind of private index marks > (additionally to index marks inserted by the client application). Ideally, > this should also somehow be handled by Festival or the other TTS in use in > the future so that Speech Dispatcher doesn't have to do anything with the > text itself. Then the whole text (in SSML) is sent to Festival for > synthesis and the function (speechd-next) Festival function is repeatedly > called to retrieve the newly synthesized text or the information about an > index mark being reached. ok > > 3) Priority models > > > KTTS Type SpeechD Type > > ---------------- --------------------- > > Text Message > > Message Important > > Warning Important > > This mapping would go strongly against the philosophy of Speech Dispatcher, > because the priority important is really only to be used for short > important messages, not for ordinary messages, to prevent ``polution'' of > the queues with important messages which can't be discarded nor postponed > by any other priority. > > Let me stop for a moment and explain our view on the message management in > accessibility. We don't imagine accessible computer for the blind as just a > screen reader, we also think about the concept of what we currently call > application reader. I'll explain the difference. > > A screen reader is a general purpose tool which you can use to do most of > your work and it should provide you a reasonable level of access to all > aplications that don't have some kind of barier in themselves. > > Screen reader typically only sees the surface of what actually is contained > in the application. It might be customized to some degree for that > particular application, but still it doesn't have access to some > information the application knows. > > So it might make sense to build accessibility to some applications > directly, so that these accessibility solutions can take advantage of the > particular details of the application and of the information that are > available to it when it operates *inside* the application. > > I'll give two examples. > > One of them is what we are currently doing with Emacs in speechd-el. > speechd-el knows the origin of each message so it can decide which > information is and which isn't important for the user (according to user > configuration of course) and take advantage of the advanced priority model. > It knows the language and character encoding of buffers, so it can switch > languages automatically. It can perform other actions that a general > purpose screen reader couldn't. Maybe Milan Zamazal can give better > examples. > > Another example is the GNU Typist package. If you are not familiar with, > it's a very good tool to help people learn touch typing. While in the > previous case of Emacs, building accessibility into the application > directly and not relying on a general purpose screen reader was a > convenience, for GNU Typist that will be an absolute need. I can't imagine > how would you be trying to learn touch typing just by reading the > application with a general purpose screen reader -- you need dictation, you > need notification about errors and so on and so on, something that a > general purpose screen reader without any understanding of the application > can't provide. > > Of course these specialized solutions, which we call application readers, > should not be hardwired into the main program. Rather, they should be some > extension or an additional layer built on top of the program. > > Let's return to the priority model. Now you understand that once we bypass > the earliest stage in the GUI, we will get to the situation that we already > are in textmode. There will typically be more clients connected to the high > level TTS API in use (KTTSD, Speech Dispatcher), not just the screen > reader. Some of them might be able to make a good use of the whole priority > system, some not. > > Now you come to understand why we don't strictly separate the screen reader > and others in Speech Dispatcher and why we don't have any priority that an > application could use to have a full and complete controll over everything > said. Rather, applications are supposed to use priorities as text and > message and when such a situation happens that there is some more important > message to be spoken (either originating in these applications or somewhere > else), than this message *should* pass through and the original messages > will be postponed or discarded. > > I think thinking in these terms should be the basis for our future work. > Obviously, the priority system in Speech Dispatcher is not optimal and we > will have to continue working on it. > > Milan explained the ideas under the different priorities in a separate > email, I tried to explain the motivation for them. I very much welcome all > your comments about all of this. I think the element that is missing is thinking *only* in the context of accessibility. Those with good sight will still want to make use of TTS for reading of ebooks, or speaking web pages while viewing something else. The problem with the current SpeechD model, as I explained, is its tendency to discard text for all but high priority messages. For an ebook reader, that's just not acceptable. > > I really like the ``long text'' idea. Maybe the nicest behavior would be if > every incomming higher-priority message would just pause reading ``long > text'' and then resume it again after the higher-priority message is > spoken. This way, a user could be reading an ebook while still being able > to listen to time and new-email notifications, being able to jump into the > mixer to adjust the sound volume etc. What do you think of it? This would solve the problem. The KTTS to SpeechD mapping might then look like this: KTTS Type SpeechD Type ---------------- --------------------- text long text warning/message message screen reader see below We will want to enhance the KTTS API to offer all the SpeechD priorities, so Screen Readers would use whatever is appropriate (important, message, text, progress, notification). (Since there are no Screen Readers in KDE at the moment, it is not a problem to remove the existing Screen Reader type from the KTTS API.) We should also deprecate the KTTS warning/message types and advise programmers to choose from the full SpeechD priority model depending upon the application's needs. > > 4) Conclusions > > > Given these issues, I cannot presently move forward with integrating > > SpeechD with KTTS. If I had my way, SpeechD would offer callbacks to > > notify when messages have been spoken. > > It will. > In summary, I have two essential needs in order to move forward with integrating KTTS with SpeechD. 1. Callbacks. (Begin/End of sentence/job. By "job", I mean that I need to know when the last sentence of a request has been spoken, otherwise, I need to do sentence boundary detection in order to count the sentences to know when a job has finished.) 2. long text priority As I mentioned, I'll send some ideas for callbacks in a separate email. I want to mention timing a bit. As you probably know, KDE will be switching to Qt4 beginning sometime in the next few months. (I believe Qt4 final is scheduled for June of this year.) Qt4 offers the GUI framework that makes a KDE Screen Reader possible. We will most likely adapt Gnopernicus to use the Qt4/AT-SPI, but that is by no means decided for sure. It would be a good thing if the KTTS/SpeechD integration were well on its way to being finished when the time comes to integrate with Qt4/AT-SPI/Gnopernicus. I can't say with certainty, but I believe early Fall would be the timeframe when all this will be occurring. It would be *really* nice if we could announce the completion (or near completion) of KTTS/SpeechD integration at the annual KDE conference (aKademy 2005), which begins August 26th. Do you think this might be possible? Regards -- Gary Cramblitt (aka PhantomsDad) KDE Text-to-Speech Maintainer http://accessibility.kde.org/developer/kttsd/index.php From garycramblitt@comcast.net Tue May 10 02:41:31 2005 From: Gary Cramblitt To: speechd-discuss@nongnu.org Date: Sun Aug 13 12:52:48 2006 Subject: [Speechd] callbacks Message-ID: <200505092041.31580.garycramblitt@comcast.net> Here's my suggestions for implementing callbacks in Speech Dispatcher. My C/C++ skills aren't perfect, so.. The traditional subscribe/callback mechanism. Define your event types, i.e., begin-sentence, end-sentence, end-job, etc. Define a bunch of constants, assigning one bit to each type of event. This allows applications to pass an event mask when they subscribe specifying the kind of events they'd like to receive. As you probably know, C libraries cannot call into C++ object methods. To work around this, allow apps to pass a void* when subscribing. Speech Dispatcher stores this along with the app's subscription information and passes it back in each callback. C++ uses this by passing an object "this" reference when subscribing and then using the passed-back pointer in the callback handler to call into the object that subscribed. Other applications can use the pointer for whatever they wish, or just pass a NULL. So a subscription call from inside C++ class called MyObj might look like this: int event_mask = SPD_EVENT_STARTSENTENCE; spd_set_callback(connection, &handle_spd_event, event_mask, this); with the callback handler coded like this: extern "C" { static int handle_spd_event(spd_event* event, int type, void* objptr); } int handle_spd_event(spd_event* event, int type, void* objptr) { return static_cast(objptr)->spdEvent(event, type); } spdEvent is a method in MyObj for handling the event. In this example, "type" will equal SD_EVENT_STARTSENTENCE. "event" is a pointer to a structure within Speech Dispatcher (or perhaps instead, an event ID) containing information about the event. The application can make calls into Speech Dispatcher, passing this pointer/ID, to request additional information about the event. For example int sentence_num = spd_get_event_sentence_num(event); The return value of the callback can be used for feedback to Speech Dispatcher. Normally, the handler should return 0. But perhaps a value of 1 would indicate to Speech Dispatcher to stop the synthesis. At any rate, you should allow a return code for future flexibility. The nice thing about this setup is you can extend it later by adding additional event types, event information retrieval routines, or handler return codes without breaking existing apps. Some other issues/ideas. You'll want to insulate Speech Dispatcher callbacks from long-running callback handlers. In KTTSD, I plan to decouple the callback from Speech Dispatcher by posting an event into the Qt event loop and returning immediately, but you might want to think about insulating Speech Dispatcher from other apps, perhaps by issuing callbacks in a separate thread. (But then, I know threads can be problematic on some platforms..) It might be handy to pass an additional parameter or two in the callback that contains most-used information, so that the handler doesn't have to call into Speech Dispatcher to retrieve event information. For example, KTTSD will want to know the "job" number and sentence number when receiving SENTENCESTARTED, so if this info were in the callback, a call into Speech Dispatcher would be unnecessary. How long should Speech Dispatcher keep event information around? I think the rule "It keeps the information until the handler returns." should be adequate. I don't know how well the void* pointer method would work for other object-oriented languages, like Python. Anyone? You'll probably want to typedef some of the "int's" in my sample code above for future expansion, i.e., the callback prototype might be sbd_result_t handle_spd_event(spd_event* event, spd_eventtype_t type, void* objptr) where sbd_result_t and spd_eventtype_t are typedef'd to int. How much information would be available for each event type? For starters, maybe not much, but you'll probably want to keep pointers to running job data structures. The problem with that is, what happens if a callback handler cancels the job on which the event was issued? Or stated another way, re-entrancy is an issue. For starters, I see no problem with limiting calls from event handlers to the "get_event_info" kind, but it is something to think about. -- Gary Cramblitt (aka PhantomsDad) KDE Text-to-Speech Maintainer http://accessibility.kde.org/developer/kttsd/index.php From garycramblitt@comcast.net Wed May 11 02:27:52 2005 From: Gary Cramblitt To: speechd-discuss@nongnu.org Date: Sun Aug 13 12:52:48 2006 Subject: [Speechd] KTTS and Sentence Boundary Detection In-Reply-To: <20050504211837.GA2246@brailcom.cz> References: <200504232319.01128.garycramblitt@comcast.net> <20050504211837.GA2246@brailcom.cz> Message-ID: <200505102027.52195.garycramblitt@comcast.net> As I mentioned in my other email, I'd like to comment on the following and point out how KTTS addressed these issues. It is academic, since the intention is to remove Sentence Boundary Detection (SBD) from KTTS, as long as the functionality needed is provided by Speech Dispatcher. It may still be of interest to you however. On Wednesday 04 May 2005 05:18 pm, Hynek Hanke wrote: > There is a number of issues we found with cutting text into sentences in > Speech Dispatcher and sending just sentences to the output modules and > synthesizers. > > 1) The TTS has no possibility to do more advanced syntactic analysis. It is > only allowed to operate on one sentence. In practice, none of the current synths alter the speaking attributes of one sentence based on surrounding sentences. It is true that syntactic analysis does assist Festival at deciding where sentence boundaries are. If we assume that KTTS puts sentence boundaries at the same places as Festival (and for the most part it does), then the end result is the same. > > 2) We need to handle language dependent issues in a project (Speech > Dispatcher, KTTSD) that should be language independent. KTTS addresses differences in SBD for different languages by using a modular plugin architecture. In theory, languages that use different punctuation and so forth for sentences can be implemented in a separate SBD filter. In practice, so far, only the Polish language has needed a separate SBD filter, mostly because Polish Festival incorrectly "speaks" punctuation characters. ("This is a sentence." is spoken as "This is a sentence period" -- in Polish of course.) Since SBD is implemented using regular expressions, the Polish SBD filter was a simple matter of changing the regular expression to remove the sentence punctuation while simultaneously breaking the input into sentences. > > 3) How to cut SSML or other markup into sentences? In KTTS, we addressed this with an eye towards the ability to advance and rewind by sentence. To achieve this, the SSML input is parsed using an XML parser.

and tags are obviously interpreted as sentence boundaries. Within text and CDATA Section nodes, the same regular expression as is used for plain text is used to decide where sentence boundaries are. Once the position of sentence boundaries is determined, each sentence is output with a complete set of SSML tags. In this way, each sentence gets a complete SSML context, so that when rewinding and advancing, no information is lost. For example, the following input SSML This is a sentence. So is this. This is spoken fast.

This is the fourth sentence.

becomes This is a sentence.So is this.This is spoken fast.This is the fourth sentence. In the case of Festival, SSML is then converted into SABLE tags using an XSLT conversion. This works out pretty well. Some of the current limitations are 1) we don't handle SSML "relative" attributes (), 2) Festival seems to have trouble with voice attributes, so we strip them out when converting to SABLE, and 3) we don't handle the tag, which isn't fully defined in the SSML spec anyway. We handle HTML by first making sure it is valid XHTML and then using XSLT to convert it to SSML. > > 4) How to cut data that are not an ordinary text (program source code, ...) Once again, the plugin filter architecture of KTTS can address this. We currently handle C/C++ text (we assume each EOL is a sentence boundary). In practice, speaking code has lots of other problems because of punctuation and "words" that are not in the lexicon, which tend to confuse Festival quite a bit, so there is a lot of work that still needs to be done on this. > > 5) It makes the output module much more complicated if good performance is > of concern. It's necessary to already have sent for synthesis the next > sentence before the previous one is spoken in the speakers so that the TTS > doesn't sit idle. Sentences of different length may cause unnecessary > delays. Since KTTS is designed for each synth to return a wav file, the synths can be kept busy working 3 or 4 sentences ahead while KTTS simultaneously outputs a sentence to the audio device. Hence, sentences of different lengths are not a problem. The first sentence begins speaking quickly because only a single sentence is sent to the synth for parsing and synthesis. In practice, the synths sit idle most of the time because KTTS builds up a queue of 3 or 4 sentences that have already been synthesized while the first sentence is still being heard on the audio device. As you know, synthesis time is much shorter compared to audio time. That said, SBD does add a small delay before the first sentence is spoken, and the larger the input, the longer the delay. SSML input adds additional delay. The ideal synth would not perform SBD on the entire input before it begins speaking the first sentence, but I don't believe Festival is so optimized. Does Speech Dispatcher do something to solve this problem? If the synth were so optimized, I imagine it would be problematic to provide advance/rewind capability within the synth. One of the things I'm looking forward to when we integrate KTTS with Speech Dispatcher is improved performance, since I know you've worked hard to address that. Thanks for listening. -- Gary Cramblitt (aka PhantomsDad) KDE Text-to-Speech Maintainer http://accessibility.kde.org/developer/kttsd/index.php From pdm@freebsoft.org Thu May 12 12:32:35 2005 From: Milan Zamazal To: speechd-discuss@nongnu.org Date: Sun Aug 13 12:52:48 2006 Subject: [Speechd] KTTS and Sentence Boundary Detection In-Reply-To: <200505102027.52195.garycramblitt@comcast.net> (Gary Cramblitt's message of "Tue, 10 May 2005 20:27:52 -0400") References: <200504232319.01128.garycramblitt@comcast.net> <20050504211837.GA2246@brailcom.cz> <200505102027.52195.garycramblitt@comcast.net> Message-ID: <87u0l8ivjw.fsf@blackbird.zamazal.org> >>>>> "GC" == Gary Cramblitt writes: GC> In practice, so far, only the Polish language has needed a GC> separate SBD filter, mostly because Polish Festival incorrectly GC> "speaks" punctuation characters. Shouldn't Polish Festival be fixed instead? GC> 1) we don't handle SSML "relative" attributes ( rate="+10">), 2) Festival seems to have trouble with voice GC> attributes, so we strip them out when converting to SABLE, and We support both relative prosody values and voice switching in the festival-freebsoft-utils SSML processor. GC> 3) we don't handle the tag, which isn't fully defined GC> in the SSML spec anyway. Neither we do, except for our own special attribute value detail="spell". GC> The ideal synth would not perform SBD on the entire input before GC> it begins speaking the first sentence, but I don't believe GC> Festival is so optimized. Does Speech Dispatcher do something GC> to solve this problem? We use festival-freebsoft-utils for this purpose. In case of plain text, Festival processes only as much of text as is needed for creating a complete utterance. In case of SSML, the document has to be first parsed with a SSML parser, but this is usually fast and definitely much faster than complete text processing. So we have no delays because of utterance chunking in large documents. See http://www.freebsoft.org/festival-freebsoft-utils for more information. Regards, Milan Zamazal -- Life. Don't talk to me about life. -- Marvin the Paranoid Android From hanke@brailcom.org Fri May 13 18:27:52 2005 From: Hynek Hanke To: speechd-discuss@nongnu.org Date: Sun Aug 13 12:52:48 2006 Subject: [Speechd] Client notification about message start/stop and reached index marks Message-ID: <20050513162752.GC2315@brailcom.cz> Hi everyone, we have already discussed on this list that it would be very desirable to let the client applications of Speech Dispatcher know when it starts to speak a certain message, when it terminates to speak a message and when index marks inside the message are reached. It's easy to get this information inside Speech Dispatcher and Festival even provides support for index marking in SSML with festival-freebsoft-utils. However, it's not clear how this information should be transmitted to the client application, because SSIP is a synchronous protocol. I think the following would be a reasonable solution. * SSIP will be extended so that the reply on the SPEAK command includes message identification number. (Technically, this would not be necessary, because it's already possible to obtain such information using the HISTORY commands. However it might not be practical to have to ask Dispatcher after every SPEAK request.) * Applications that want to make use of the notification open a second (synchronous) TCP/IP connection to Speech Dispatcher on which Speech Dispatcher is the sender and application is the listener. The protocol used for this would be very similar to SSIP or become a part of SSIP. I can think of 4 or 6 types of events: 1) started to speak message "msg-id" 2) index mark "mark-id" reached in message "msg-id" 3) end of speaking message "msg-id" 4) message "msg-id" discarded (canceled) 5) message "msg-id" paused (would this be useful for something?) 6) message "msg-id" resumed (would this be useful for something?) * SSIP will be extended so that the application can specify which types of events it wants to be notified about. These would be state switches valid for all incomming messages until the notification request is switched off or to some other mode. (In a similar way to how priority switching is handled now in Speech Dispatcher.) I think receiving all notification about all messages is unnecessary most of the time, it would only cause more CPU load on the side of both server and application, and the application would typically receive a lot of useless messages/callbacks (for example type (1), (3) and (4) for messages of priority PROGRESS etc.) For this reason, I think it's good to have a possibility in SSIP to choose for which messages the application is interested in receiving these notifications. Just to give an example, it could be something like SET self NOTIFICATION none SET self NOTIFICATION begin-end-discarded SET self NOTIFICATION index-marks-begin-end-discarded SET self NOTIFICATION all of course with values of a diferent name. An SSIP session on the main connection might then look like this SET self NOTIFICATION all 2??-OK NOTIFICATION SET SPEAK Hello world! How are you? . 2??-213 2??-OK MESSAGE QUEUED SET self NOTIFICATION none 2??-OK NOTIFICATION SET SPEAK blabla . 2??-213 2??-OK MESSAGE QUEUED * The language-specific APIs would offer a function that listens on the socket and returns on each received event with the full information about the event. * The language-specific APIs would offer a function that (maybe run from a separate thread) would execute a user-specified callback on receiving each event. This should all be quite easy to implement. Question: How to ensure that both connections (main and notification) are connected to the same client application? Each client application could, on request, receive a hash-code through the main connection and then use it to open the notification connection. Would that cause some security troubles? (Of course, each program that would listen to the communication between client and server on the main connection would know the hash-code and could use it for false authorization. But would that be a problem? The program wouldn't get any more sensitive data than what it already gets from listening to the communication on the main connection -- which already would be very severe.) Please let me know what do you think about it. I'd especially like to know if this is suitable from the client application point of view or if some different approach would be better (which one?) as I don't have much experience with the application side. With Regards, Hynek Hanke From garycramblitt@comcast.net Sat May 14 00:39:09 2005 From: Gary Cramblitt To: speechd-discuss@nongnu.org Date: Sun Aug 13 12:52:48 2006 Subject: [Speechd] Re: KTTSD and Speech Dispatcher In-Reply-To: <20050513112553.GA2365@brailcom.cz> References: <20050320183223.GH2197@brailcom.cz> <200505031741.26812.garycramblitt@comcast.net> <20050513112553.GA2365@brailcom.cz> Message-ID: <200505131839.10016.garycramblitt@comcast.net> On Friday 13 May 2005 07:25 am, Hynek Hanke wrote: > On Tue, May 03, 2005 at 05:41:26PM -0400, Gary Cramblitt wrote: > > I forwarded a copy of the draft audio framework (strongly indicating that > > it is draft) to the kde-multimedia mailing list. A flame war immediately > > broke out between GStreamer and NMM guy and no feedback or comments at > > all on accessibility needs.. sigh. I fear the KDE folks have thrown the > > baby out with the bath water in the premature decision to abandon aRts. > > Thank you. I've seen the follow-up discussion since I'm on the list myself. > I'm a little puzzled about how the decision about the future of KDE > Multimedia will be made and who is the head in KDE in this area? Is there > someone who could confirm our accessibility requirements will be regarded > or state why they won't and what can we change about them so that they are? AFAIK, there is no "official", or even "unofficial" lead person or coordinator. This too concerns me. If you like, I can ask on the mailing list? Maybe the question will stimulate some discussion. > > > If it isn't *too* hard to do, I think direct ALSA is our best bet right > > now. I have a *mild* interest in maybe helping out with this, if nobody > > else wants to undertake it. > > It's not very hard for someone who know how the ALSA API and ALSA itself > works. I've already begin implementing and testing ALSA output in Speech > Dispatcher, so I was waiting till I bring it to reasonable state so you > could look into it and hopefully improve it. Any help is highly welcome. Well, I have never programmed ALSA, so you are way ahead of me. I just have a mild interest in doing it. It might be useful in the future. I'll take a look at the code. It also occurred to me to look at some of the ALSA implementations in kdemultimedia and compare to what you've implemented. Maybe there are some solutions to the problems you mention. If I understood you, I can get the Speech Dispatcher ALSA code from CVS right now? > > BTW, I have a Jabber account now PhantomsDad@njs.netlab.cz. Hope to hear > > from you online. > > I just added you to my roster. Maybe I'll see you online this weekend. :-0 > > > I spoke for awhile with Jan Buchal, who wants me to go to the Libre > > Software Meeting. I told him I *hate* to travel by air, so don't count > > on my attending. > > I still think it would be very important and a wonderful oportunity to > discuss many things and move forward. Did you consider other means of > transport (I don't know where you live, so that might not be possible of > course :) ? Crofton, Maryland, USA. I'm already thinking of *not* going to aKademy 2005 in August for the same reasons, and sorry, Libre is second on my priority list. Regards -- Gary Cramblitt (aka PhantomsDad) KDE Text-to-Speech Maintainer http://accessibility.kde.org/developer/kttsd/index.php From pdm@freebsoft.org Sat May 14 18:52:53 2005 From: Milan Zamazal To: speechd-discuss@nongnu.org Date: Sun Aug 13 12:52:48 2006 Subject: [Speechd] Client notification about message start/stop and reached index marks In-Reply-To: <20050513162752.GC2315@brailcom.cz> (Hynek Hanke's message of "Fri, 13 May 2005 18:27:52 +0200") References: <20050513162752.GC2315@brailcom.cz> Message-ID: <87hdh5iwbe.fsf@blackbird.zamazal.org> >>>>> "HH" == Hynek Hanke writes: HH> * Applications that want to make use of the notification open a HH> second (synchronous) TCP/IP connection to Speech Dispatcher on HH> which Speech Dispatcher is the sender and application is the HH> listener. The protocol used for this would be very similar to HH> SSIP or become a part of SSIP. I'd recommend to use a single connection. Notifications can be simply reported in the general command answer format, with a special numeric code, to the standard SSIP connection. If they are undesirable, they can be ignored or switched off. HH> Just to give an example, it could be something like SET self HH> NOTIFICATION none SET self NOTIFICATION begin-end-discarded SET HH> self NOTIFICATION index-marks-begin-end-discarded SET self HH> NOTIFICATION all of course with values of a diferent name. I'd prefer having a single on/off switch for each type of the notification, i.e. something like SET self NOTIFICATION begin ON SET self NOTIFICATION begin OFF SET self NOTIFICATION discarded ON etc. Adding a special value `all' for enabling/disabling all kinds of notifications would be probably useful: SET self NOTIFICATION all ON SET self NOTIFICATION all OFF At the beginning of the session, all notifications should be set off, to retain compatibility with the current SSIP protocol. HH> * The language-specific APIs would offer a function that listens HH> on the socket and returns on each received event with the full HH> information about the event. I think this is not much important ... HH> * The language-specific APIs would offer a function that (maybe HH> run from a separate thread) would execute a user-specified HH> callback on receiving each event. ... while this is important. HH> Question: How to ensure that both connections (main and HH> notification) are connected to the same client application? [...] HH> Would that cause some security troubles? With a single connection, there should be no such problem. HH> Please let me know what do you think about it. With the exceptions mentioned above, I think the proposal is reasonable. HH> I'd especially like to know if this is suitable from the client HH> application point of view or if some different approach would be HH> better (which one?) as I don't have much experience with the HH> application side. With the modifications I suggest above, I think it could work well with speechd-el. Regards, Milan Zamazal -- Life. Don't talk to me about life. -- Marvin the Paranoid Android From garycramblitt@comcast.net Fri May 20 00:53:53 2005 From: Gary Cramblitt To: speechd-discuss@nongnu.org Date: Sun Aug 13 12:52:48 2006 Subject: Festival 1.95 beta Message-ID: <200505191853.53619.garycramblitt@comcast.net> Does festival-freebsoft-utils 0.4 work with Festival 1.95 beta? The README says Festival 1.4.3 and "other versions may or may not work". Same question for speechd 0.5 -- Gary Cramblitt (aka PhantomsDad) KDE Text-to-Speech Maintainer http://accessibility.kde.org/developer/kttsd/index.php From garycramblitt@comcast.net Fri May 20 04:19:20 2005 From: Gary Cramblitt To: speechd-discuss@nongnu.org Date: Sun Aug 13 12:52:48 2006 Subject: [Speechd] Client notification about message start/stop and reached index marks In-Reply-To: <20050513162752.GC2315@brailcom.cz> References: <20050513162752.GC2315@brailcom.cz> Message-ID: <200505192219.20050.garycramblitt@comcast.net> On Friday 13 May 2005 12:27 pm, Hynek Hanke wrote: > Please let me know what do you think about it. I'd especially like to know > if this is suitable from the client application point of view or if some > different approach would be better (which one?) as I don't have much > experience with the application side. I apologize for this late reply. For the most part, what you propose is good. I'm a bit concerned about the security aspects. It should be possible for a GUI used to control Speech Dispatcher in realtime to obtain events for all applications speaking on the same machine, and to control output from those applications. (In this case, the pause and resume events become useful because the controlling GUI needs to know if an application has paused/resumed its output.) On the other hand, concerns about security are an issue too. I wonder if there is some way we can satisfy both needs? Perhaps the ability to designate a single "privileged" application in the configuration file? -- Gary Cramblitt (aka PhantomsDad) KDE Text-to-Speech Maintainer http://accessibility.kde.org/developer/kttsd/index.php From pdm@freebsoft.org Fri May 20 13:55:54 2005 From: Milan Zamazal To: speechd-discuss@nongnu.org Date: Sun Aug 13 12:52:48 2006 Subject: Festival 1.95 beta In-Reply-To: <200505191853.53619.garycramblitt@comcast.net> (Gary Cramblitt's message of "Thu, 19 May 2005 18:53:53 -0400") References: <200505191853.53619.garycramblitt@comcast.net> Message-ID: <8764xem7qt.fsf@blackbird.zamazal.org> >>>>> "GC" == Gary Cramblitt writes: GC> Does festival-freebsoft-utils 0.4 work with Festival 1.95 beta? [...] GC> Same question for speechd 0.5 Yes and yes. Regards, Milan Zamazal -- As is typical in civilized (i.e., non-corporate) email discussions, my remarks *follow* (not precede) their context. -- Branden Robinson in debian-legal From garycramblitt@comcast.net Sun Jun 5 16:23:15 2005 From: Gary Cramblitt To: speechd-discuss@nongnu.org Date: Sun Aug 13 12:52:48 2006 Subject: stderr? Message-ID: <200506051023.15663.garycramblitt@comcast.net> This is probably a really stupid question, but I'm out of ideas.. I'm hacking a bit on the new alsa code, inserting debugging statements using fprintf(stderr, ...), but I don't see them displayed anywhere. I've looked in speechd.log. I'm running speech-dispatcher in Konsole A like this speech-dispatcher -s -l 5 and running spd-say in Konsole B spd-say "Hello World" but my debugs don't display in either Konsole. Also, I could use some tips on the best way to work on Speech Dispatcher code. ATM, I have to install as root each time I make a change, i.e., "make install". How should I set up my environment so I don't have to do that? -- Gary Cramblitt (aka PhantomsDad) KDE Text-to-Speech Maintainer http://accessibility.kde.org/developer/kttsd/index.php From garycramblitt@comcast.net Sun Jun 5 22:22:20 2005 From: Gary Cramblitt To: speechd-discuss@nongnu.org Date: Sun Aug 13 12:52:48 2006 Subject: stderr? In-Reply-To: <200506051023.15663.garycramblitt@comcast.net> References: <200506051023.15663.garycramblitt@comcast.net> Message-ID: <200506051622.20994.garycramblitt@comcast.net> On Sunday 05 June 2005 10:23 am, Gary Cramblitt wrote: > This is probably a really stupid question, but I'm out of ideas.. > > I'm hacking a bit on the new alsa code, inserting debugging statements > using fprintf(stderr, ...), but I don't see them displayed anywhere. OK, I figured out I must run sd_festival separately to see debugs from audio routines. > Also, I could use some tips on the best way to work on Speech Dispatcher > code. ATM, I have to install as root each time I make a change, i.e., "make > install". How should I set up my environment so I don't have to do that? Still hoping for tips. -- Gary Cramblitt (aka PhantomsDad) KDE Text-to-Speech Maintainer http://accessibility.kde.org/developer/kttsd/index.php From aab@cichlid.com Sun Jun 5 23:01:11 2005 From: Andrew Burgess To: speechd-discuss@nongnu.org Date: Sun Aug 13 12:52:48 2006 Subject: stderr? In-Reply-To: <200506051023.15663.garycramblitt@comcast.net> References: <200506051023.15663.garycramblitt@comcast.net> Message-ID: On Sun, 05 Jun 2005 07:23:15 -0700, Gary Cramblitt wrote: > This is probably a really stupid question, but I'm out of ideas.. This is probably a stupid answer but do you: fflush(stderr); ever? otherwise they are buffered up and you don't see lines right away... > > I'm hacking a bit on the new alsa code, inserting debugging statements > using > fprintf(stderr, ...), but I don't see them displayed anywhere. From pdm@freebsoft.org Mon Jun 6 04:12:11 2005 From: Milan Zamazal To: speechd-discuss@nongnu.org Date: Sun Aug 13 12:52:48 2006 Subject: stderr? In-Reply-To: <200506051023.15663.garycramblitt@comcast.net> (Gary Cramblitt's message of "Sun, 5 Jun 2005 10:23:15 -0400") References: <200506051023.15663.garycramblitt@comcast.net> Message-ID: <877jh86xok.fsf@blackbird.zamazal.org> >>>>> "GC" == Gary Cramblitt writes: GC> Also, I could use some tips on the best way to work on Speech GC> Dispatcher code. ATM, I have to install as root each time I GC> make a change, i.e., "make install". How should I set up my GC> environment so I don't have to do that? Provide suitable prefix and pidpath to configure, e.g.: ./configure --prefix=$HOME/tmp/speech-dispatcher pidpath=$HOME/tmp Then you can install and run Speech Dispatcher as a non-root user, provided the user has appropriate access permissions to the audio hardware. Regards, Milan Zamazal -- The seeker after truth should be humbler than the dust. The world crushes the dust under its feet, but the seeker after truth should so humble himself that even the dust could crush him. -- M. K. Gandhi From hanke@brailcom.org Tue Jun 7 18:25:18 2005 From: Hynek Hanke To: speechd-discuss@nongnu.org Date: Sun Aug 13 12:52:48 2006 Subject: stderr? In-Reply-To: <200506051023.15663.garycramblitt@comcast.net> References: <200506051023.15663.garycramblitt@comcast.net> Message-ID: <20050607162518.GB2462@brailcom.cz> On Sun, Jun 05, 2005 at 10:23:15AM -0400, Gary Cramblitt wrote: > I'm hacking a bit on the new alsa code, inserting debugging statements using > fprintf(stderr, ...), but I don't see them displayed anywhere. I've looked > in speechd.log. I'm running speech-dispatcher in Konsole A like this > speech-dispatcher -s -l 5 They must appear at the console you are running speech-dispatcher from. I suspect this is a buffering issue. Just use fflush(). > Also, I could use some tips on the best way to work on Speech Dispatcher code. > ATM, I have to install as root each time I make a change, i.e., "make > install". How should I set up my environment so I don't have to do that? As Milan pointed out, you don't have to install as root. If you are working on the audio output library or some language specific API, you do have to ,,make install'' so that the changes have some effect. You don't have to do this when you are working on the core or when working with output modules (in this case you have to set up your /etc/speech-dispatcher/speechd.conf to correct paths into your build directory). With Regards, Hynek Hanke From hanke@brailcom.org Tue Jun 7 18:29:16 2005 From: Hynek Hanke To: speechd-discuss@nongnu.org Date: Sun Aug 13 12:52:48 2006 Subject: stderr? In-Reply-To: <200506051622.20994.garycramblitt@comcast.net> References: <200506051023.15663.garycramblitt@comcast.net> <200506051622.20994.garycramblitt@comcast.net> Message-ID: <20050607162916.GC2462@brailcom.cz> > OK, I figured out I must run sd_festival separately to see debugs from audio > routines. You should not have to. sd_festival is run as a child process by speech-dispatcher. The stderr should not be affected by this. Of course it would be best if it would be possible to redirect stderr from child processes to some logfile, but I don't think this is possible. With Regards, Hynek Hanke From pollock.nageoire@wanadoo.fr Thu Jun 16 12:01:58 2005 From: Pierre Lorenzon To: speechd-discuss@nongnu.org Date: Sun Aug 13 12:52:48 2006 Subject: speech dispatcher en french. Message-ID: <20050616.120158.74748735.pollock.nageoire@wanadoo.fr> Hi, I discovered recently the speech dispatcher and I am very interested in it. Indeed my goal was to make an interface between emacspeak and festival. Festival server features were to limited for a complete implementation of all emacspeak requirements. I started to developed a direct client in emacspeak for the festival server but there were many things to be added to get all emacspeak functionalities working. But speech dispatcher offers the ideal solution since all what I wanted to implement was already in freebsoft utils. And as we say in French "je ne veux pas r?inventer la roue." There is now "only one thing to do" i.e. an emacspeak client for the speech dispatcher. Even if it is said in the speechd-el documentation that it does not seem to be an easy work I know enough emacspeak code to do that. To sum up there is only a new xxx-interp.el module to be implemented an I think that the speechd-speak.el might be very helpful. I know that the emacspeak implementation is not modular enough and all sorts of thing but I am an emacspeak user and appreciate the concept. But first, since I am a french native speaker I'd be very pleased to let the system emacspeak | speech-dispatcher | festival |`(mbrola) speak French. We have a French customization of festival called FranFest which works with festival so French is available in my festival. What should I do to let freebsoft utils (and the speech dispatcher) know that French is available ? In other words, what should I do in order that the (speechd-set-language "fr") takes effect ? I added an entry in the language-codes variable (fr french) but it is not enough. Bests Pierre From pdm@freebsoft.org Thu Jun 16 21:40:43 2005 From: Milan Zamazal To: speechd-discuss@nongnu.org Date: Sun Aug 13 12:52:48 2006 Subject: speech dispatcher en french. In-Reply-To: <20050616.120158.74748735.pollock.nageoire@wanadoo.fr> (Pierre Lorenzon's message of "Thu, 16 Jun 2005 12:01:58 +0200 (CEST)") References: <20050616.120158.74748735.pollock.nageoire@wanadoo.fr> Message-ID: <87fyviay4k.fsf@blackbird.zamazal.org> >>>>> "PL" == Pierre Lorenzon writes: PL> What should I do to let freebsoft utils (and the speech PL> dispatcher) know that French is available ? In other words, what PL> should I do in order that the (speechd-set-language "fr") takes PL> effect ? I added an entry in the language-codes variable (fr PL> french) but it is not enough. In order festival-freebsoft-utils can find the French voice, the voice must be properly proclaimed and the `language' property must be set properly. Check for a `proclaim_voice' call in FranFest. Additionally, setting the special `coding' voice property introduced by festival-freebsoft-utils is necessary if the voice works with an input text character coding other than ISO-8859-1. FWIW, here's some proclaim_voice example from the Czech Festival support (it should be intelligible to Lisp programmers:-): (defmac (czech-proclaim-voice form) (let ((name (nth 1 form)) (description (nth 2 form)) (body (nth_cdr 3 form)) (options ())) (if (consp name) (begin (set! options (cdr name)) (set! name (car name)))) (set! name (intern (string-append 'czech_ name))) (let ((parameters `((language czech) (dialect ,(cdr (assoc 'dialect options))) (gender ,(cadr (assoc 'gender options))) (coding ISO-8859-2) (description ,description)))) `(begin (define (,(intern (string-append 'voice_ name))) (czech-reset-parameters) ,@body (voice-czech-common) (set! current-voice (quote ,name))) (proclaim_voice (quote ,name) (quote ,parameters)))))) Regards, Milan Zamazal -- Wasting somebody else's time strikes me as the height of rudeness. Bill Gates From pollock.nageoire@wanadoo.fr Thu Jun 16 23:04:23 2005 From: Pierre Lorenzon To: speechd-discuss@nongnu.org Date: Sun Aug 13 12:52:48 2006 Subject: speech dispatcher en french. In-Reply-To: <87fyviay4k.fsf@blackbird.zamazal.org> References: <20050616.120158.74748735.pollock.nageoire@wanadoo.fr> <87fyviay4k.fsf@blackbird.zamazal.org> Message-ID: <20050616.230423.74740417.pollock.nageoire@wanadoo.fr> Hi Thanks for your messagge. I already started with the emacspeak client for speech dispatcher. It is not fully functional but it works ! Bests Pierre From pollock.nageoire@wanadoo.fr Fri Jun 17 11:15:25 2005 From: Pierre Lorenzon To: speechd-discuss@nongnu.org Date: Sun Aug 13 12:52:48 2006 Subject: More about speechd-el In-Reply-To: <87fyviay4k.fsf@blackbird.zamazal.org> References: <20050616.120158.74748735.pollock.nageoire@wanadoo.fr> <87fyviay4k.fsf@blackbird.zamazal.org> Message-ID: <20050617.111525.74722876.pollock.nageoire@wanadoo.fr> Hi I tried speechd-el and definitely I like it ! I am sure that I will be able to interface it with ``upper modules'' of emacspeak that I like too. But for the moment I just experiment. I noticed a strange behaviors of speechd-el : Is there a problem with parenthesis ? When a sentence is inclosed in parenthesis it is sometimes not pronounced. Moreover the parenthesis as isolated character doesn"'t seem to be pronounced as well. Is it known problem ? Might it come from my festival even if I didn't see anything suspicious in the log files. Bests Pierre From pollock.nageoire@wanadoo.fr Fri Jun 17 16:48:53 2005 From: Pierre Lorenzon To: speechd-discuss@nongnu.org Date: Sun Aug 13 12:52:48 2006 Subject: Parenthesis In-Reply-To: <87fyviay4k.fsf@blackbird.zamazal.org> References: <20050616.120158.74748735.pollock.nageoire@wanadoo.fr> <87fyviay4k.fsf@blackbird.zamazal.org> Message-ID: <20050617.164853.74727491.pollock.nageoire@wanadoo.fr> Hi, The problem with parenthesis occurs only in all punctuation mode and when trying to spell text, i.e; when the system tries to pronounce the parenthesis. I discovered that it seems to be a server-side problem. Indeed when I try to use speech-dispatcher in festival interactive mode and execute a command like : (speechd-speak* "(hello)") I get following error message in festival : Cannot concatenate waveforms with differing numbers of channels. The problem doesn't appea with other punctuations like [ ] or whatever. Bests Pierre From pdm@freebsoft.org Fri Jun 17 20:47:17 2005 From: Milan Zamazal To: speechd-discuss@nongnu.org Date: Sun Aug 13 12:52:48 2006 Subject: Parenthesis In-Reply-To: <20050617.164853.74727491.pollock.nageoire@wanadoo.fr> (Pierre Lorenzon's message of "Fri, 17 Jun 2005 16:48:53 +0200 (CEST)") References: <20050616.120158.74748735.pollock.nageoire@wanadoo.fr> <87fyviay4k.fsf@blackbird.zamazal.org> <20050617.164853.74727491.pollock.nageoire@wanadoo.fr> Message-ID: <8764wcbz2i.fsf@blackbird.zamazal.org> >>>>> "PL" == Pierre Lorenzon writes: PL> The problem with parenthesis occurs only in all punctuation mode PL> and when trying to spell text, i.e; when the system tries to PL> pronounce the parenthesis. I discovered that it seems to be a PL> server-side problem. Indeed when I try to use speech-dispatcher PL> in festival interactive mode and execute a command like : PL> (speechd-speak* "(hello)") I get following error message in PL> festival : Cannot concatenate waveforms with differing numbers PL> of channels. The problem doesn't appea with other punctuations PL> like [ ] or whatever. See the Festival variable word-mapping. It maps parentheses to sounds by default. You can either set the variable to nil in your Festival startup file or (which may be more interesting) you can install the sound-icons package (available at http://www.freebsoft.org/pub/projects/sound-icons/ or as the `sound-icons' Debian package). Regards, Milan Zamazal -- I fear fear both faith and power They have no need to use any question marks -- UJD From pollock.nageoire@wanadoo.fr Fri Jun 17 21:38:11 2005 From: Pierre Lorenzon To: speechd-discuss@nongnu.org Date: Sun Aug 13 12:52:48 2006 Subject: Parenthesis In-Reply-To: <8764wcbz2i.fsf@blackbird.zamazal.org> References: <87fyviay4k.fsf@blackbird.zamazal.org> <20050617.164853.74727491.pollock.nageoire@wanadoo.fr> <8764wcbz2i.fsf@blackbird.zamazal.org> Message-ID: <20050617.213811.104053610.pollock.nageoire@wanadoo.fr> Thanks again I try now to make speech-dispatcher speak french. I simply downloaded the czech festival package and will take it as a model since the greatest part of the work is already done in FranFest and I simply need scheme routines. Bests Pierre From pollock.nageoire@wanadoo.fr Sat Jun 18 09:24:31 2005 From: Pierre Lorenzon To: speechd-discuss@nongnu.org Date: Sun Aug 13 12:52:48 2006 Subject: FranFest works with the speech dispatcher Message-ID: <20050618.092431.74723561.pollock.nageoire@wanadoo.fr> Hi, Now problem now, Franfest works with the speech dispatcher as well as with the emacs client. The puntuation mode doesn't works so good as in english but it comes from the fact that certain symbols are not pronounced by FranFest and will be improved soon. Regards Pierre From pollock.nageoire@wanadoo.fr Tue Jun 21 18:42:35 2005 From: Pierre Lorenzon To: speechd-discuss@nongnu.org Date: Sun Aug 13 12:52:48 2006 Subject: Strange coding system behavior Message-ID: <20050621.184235.71090287.pollock.nageoire@wanadoo.fr> Hi, The use of speedh-el induces a strange behavior of the conding system which is a problem when trying to compile a latex file with iso latin 1 inputenc. Indeed the buffer-file-coding-system is changed to raw-text even if the file has been loaded with iso-latin-1. After that latex cannot compile this file. When I try the same operations with emacs only without speechd-el the coding system change does not occur. I have set the language option to french in emacs but it does not have any effect for this point. Bests Pierre From pdm@freebsoft.org Tue Jun 21 20:22:44 2005 From: Milan Zamazal To: speechd-discuss@nongnu.org Date: Sun Aug 13 12:52:48 2006 Subject: Strange coding system behavior In-Reply-To: <20050621.184235.71090287.pollock.nageoire@wanadoo.fr> (Pierre Lorenzon's message of "Tue, 21 Jun 2005 18:42:35 +0200 (CEST)") References: <20050621.184235.71090287.pollock.nageoire@wanadoo.fr> Message-ID: <873brbvabv.fsf@blackbird.zamazal.org> >>>>> "PL" == Pierre Lorenzon writes: PL> The use of speedh-el induces a strange behavior of the conding PL> system which is a problem when trying to compile a latex file PL> with iso latin 1 inputenc. Indeed the buffer-file-coding-system PL> is changed to raw-text even if the file has been loaded with PL> iso-latin-1. After that latex cannot compile this file. When I PL> try the same operations with emacs only without speechd-el the PL> coding system change does not occur. Please try this patch, I think it should fix the bug: -------------- next part -------------- A non-text attachment was scrubbed... Name: speechd-el.patch Type: text/x-patch Size: 508 bytes Desc: not available Url : http://lists.freebsoft.org/pipermail/speechd/attachments/20050621/89da3eb7/speechd-el.bin -------------- next part -------------- Regards, Milan Zamazal -- Free software is about freedom, not about free beer. If you care only about the latter, you'll end up with no freedom and no free beer. From pollock.nageoire@wanadoo.fr Tue Jun 21 20:44:06 2005 From: Pierre Lorenzon To: speechd-discuss@nongnu.org Date: Sun Aug 13 12:52:48 2006 Subject: Strange coding system behavior In-Reply-To: <873brbvabv.fsf@blackbird.zamazal.org> References: <20050621.184235.71090287.pollock.nageoire@wanadoo.fr> <873brbvabv.fsf@blackbird.zamazal.org> Message-ID: <20050621.204406.74725265.pollock.nageoire@wanadoo.fr> Hi, Thanks indeed it solves the problem. Bests Pierre From pollock.nageoire@wanadoo.fr Wed Jun 22 14:27:31 2005 From: Pierre Lorenzon To: speechd-discuss@nongnu.org Date: Sun Aug 13 12:52:48 2006 Subject: Setting language. Message-ID: <20050622.142731.74747210.pollock.nageoire@wanadoo.fr> Hi, I noticed that the set language mechanism is not similar to the other set parameter mechanisms. The method speechd-set-language is not build by the speechd--generate-set-command macro. Moreover the method speechd-set-language does not match its documentation in the sens that the documetation claims that the set-language is a local action and the implementation sets a global value. I personnaly think that the language setting should be a local operation (i.e. proper to one connection and maybe one buffer). Indded if you set a new language for a buffer you not necessarily need that all messages are affected by this setting. Hence I removed the definition of speechd-set-language and let it defined by the speechd--generate-set-command macro. It does not seem to generate any problem for the moment. Were their good reasons that I did not notice not to implement set-language with the speechd--generate-set-command macro ? Bests Pierre From pdm@freebsoft.org Wed Jun 22 17:35:38 2005 From: Milan Zamazal To: speechd-discuss@nongnu.org Date: Sun Aug 13 12:52:48 2006 Subject: Setting language. In-Reply-To: <20050622.142731.74747210.pollock.nageoire@wanadoo.fr> (Pierre Lorenzon's message of "Wed, 22 Jun 2005 14:27:31 +0200 (CEST)") References: <20050622.142731.74747210.pollock.nageoire@wanadoo.fr> Message-ID: <87k6kmqu9h.fsf@blackbird.zamazal.org> >>>>> "PL" == Pierre Lorenzon writes: PL> the method speechd-set-language does not match its documentation PL> in the sens that the documetation claims that the set-language PL> is a local action and the implementation sets a global value. Strictly speaking speechd-set-language does what the docstring says, but additionally it sets the speechd-language variable. I think it shouldn't set it, it should be done in speechd-speak.el. I'll fix it. [...] PL> Hence I removed the definition of speechd-set-language and let PL> it defined by the speechd--generate-set-command macro. It does PL> not seem to generate any problem for the moment. It shouldn't cause any problem if you use just the speechd.el library. PL> Were their good reasons that I did not notice not to implement PL> set-language with the speechd--generate-set-command macro ? I think so, but I can't recall them. Perhaps the non-standard docstring? Regards, Milan Zamazal -- The rush to reproduce Microsofts window environment seems to overshadow the design process of determining what a window environment should be, and what its ultimate users will want. -- Barry Fishman in gnu.misc.discuss From cerha@brailcom.org Tue Oct 11 01:40:55 2005 From: Tomas Cerha To: speechd-discuss@nongnu.org Date: Sun Aug 13 12:52:48 2006 Subject: About speech dispatcher In-Reply-To: <20050928120541.GA7421@iris> References: <20050928120541.GA7421@iris> Message-ID: <434AFC07.6090000@brailcom.org> Yannick PLASSIARD wrote: > Hello: > I'm interrested in this project and would like to develop an > interface to a French synthesis system callec `cicero`. This > TTS is written in Python. Hello Yannick, Sorry for a late reply. I just got back from a vacation... The key developer of Speech Dispetcher is Hynek Hanke. I CC this message to him, so he will probably give you more information. I also CC to the Speech Dispatcher development mailing-list. Writing an output module for Cicero should not be a problem. Technically, the module could be even written in Python, since Speech Dispatcher's output modules run as separate processes communicating with the core over a socket, but I'm not sure whether there is not some limitation imposed by server implementation (loading the plugins etc.). I guess Hynek will comment on that. > Are there any Python interfaces to speech-dispatcher? There is currently a Python client library. I'm not sure how complete it is, but in general it allows you to send messages directly from your application to Speech Dispatcher. This will not help you with writing the output module, but it could be extended in such a way, since the communication protocol of output modules and the server core uses very simillar rules as SSIP, which is the client protocol implemented by the Python client library. > Is there a CVS or SVN repository to get the last developpment release > of speech-dispatcher? You can find information about Speech Dispatcher CVS at http://www.freebsoft.org/speechd. There is also a complete developers documentation (including information about writing output modules). > Is the project always maintained (I saw your name on > www.freebsoft.org/). Yes, the project is still in active development and I recommend subscribing to the mailing list at http://lists.freebsoft.org/mailman/listinfo/speechd. > Moreover, a second project would be to make a link bitween > gnome-speech and speech-dispatcher; what do you think about this idea ? Yes, we cooperate with Gnome people on that as well as with KDE people on KTTS and Speech Dispatcher integration. I'm not sure what was the latest development in this area. Would someone else add this information? Kindest regards Tomas Cerha -- Brailcom, o.p.s. http://www.brailcom.org Free(b)soft project http://www.freebsoft.org Eurochance project http://eurochance.brailcom.org From nath.ml@free.fr Tue Nov 8 21:25:53 2005 From: Nath To: speechd-discuss@nongnu.org Date: Sun Aug 13 12:52:49 2006 Subject: speech-dispatcher + DECtalk on Debian AMD x86-64 : no sound In-Reply-To: <86r79rxylt.fsf@brailcom.org> (Jan Buchal's message of "Tue, 08 Nov 2005 10:44:46 +0100") References: <87y840umfx.fsf@free.fr> <86r79rxylt.fsf@brailcom.org> Message-ID: <87fyq63mzy.fsf@free.fr> Jan Buchal writes: > Bonjour, > >>>>>> "N" == Nath writes: > [.....] > > > N> do you have an idea of what can cause the problem ? > > Not yet. Maybe in configuration. Can you please send you speechd.conf? > > have you nice day > > -- > > Jan Buchal > Tel: (00420) 24 24 86 008 > Mob: (00420) 608023021 > > _______________________________________________ > Speechd mailing list > Speechd@lists.freebsoft.org > http://lists.freebsoft.org/mailman/listinfo/speechd > > Hello ! In fact I modified a little my speechd.conf file because it appeared that the modules festival and flite were not commented and then speech-dispatcher tried to load them. so now only dtk-generic is specified and there is no more errors in the /var/log/speech-dispatcher/speech-dispatcher.log file when I try to use spd-say. But unfortunatly always no sound at all ! For info my soundcard works well since I can hear mp3s without trouble and the DECtalk say program works too when I run it separatly from speech-dispatcher. As you ask me, here is the speechd.conf file (see attachment) Thx in advance for help -------------- next part -------------- A non-text attachment was scrubbed... Name: speechd.conf Type: application/octet-stream Size: 6007 bytes Desc: not available Url : http://lists.freebsoft.org/pipermail/speechd/attachments/20051108/c111c29a/speechd.obj -------------- next part -------------- -- Nath -------------- next part -------------- _______________________________________________ Speechd mailing list Speechd@lists.freebsoft.org http://lists.freebsoft.org/mailman/listinfo/speechd From pollock.nageoire@wanadoo.fr Thu Dec 22 00:13:22 2005 From: Pierre Lorenzon To: speechd-discuss@nongnu.org Date: Sun Aug 13 12:52:49 2006 Subject: Speech-dispatcher initialization In-Reply-To: <1133946192.4033.8.camel@chopin> References: <20051207.091350.102550073.pollock.nageoire@wanadoo.fr> <1133946192.4033.8.camel@chopin> Message-ID: <20051222.001322.112623657.pollock.nageoire@wanadoo.fr> Hi, We try to start speech-dispatcher again (v 0.5). The installation of the whole system was completely rebuilt avoiding problems with multiple installations of speech-dispatcher. It now works but partially. We set festival as default module and as module for english and french and used the speechd-el client in conjunction with emacs as client. The first think I noticed and which seemms to me to be unusual, is that when we switch to the french language inside emacs, a second sd_festival is created. On another installation where I usually run speech-dispatcher there is always only one sd_festival. The two sd_festival can live together a few moment but one of them soon dies and then there no longer sound. I enabled the debug mode in the festival.conf file and got a debug file which seems to be suspicious. Here is this file is it normal ? Best regards Pierre -------------- next part -------------- A non-text attachment was scrubbed... Name: debug-festival Type: application/octet-stream Size: 147756 bytes Desc: not available Url : http://lists.freebsoft.org/pipermail/speechd/attachments/20051222/52ea1f0e/debug-festival.obj -------------- next part -------------- _______________________________________________ Speechd mailing list Speechd@lists.freebsoft.org http://lists.freebsoft.org/mailman/listinfo/speechd From buchal@brailcom.org Tue Nov 8 10:44:46 2005 From: Jan Buchal To: speechd-discuss@nongnu.org Date: Sun Aug 13 12:52:49 2006 Subject: speech-dispatcher + DECtalk on Debian AMD x86-64 : no sound In-Reply-To: <87y840umfx.fsf@free.fr> (Nath's message of "Mon, 07 Nov 2005 23:23:30 +0100") References: <87y840umfx.fsf@free.fr> Message-ID: <86r79rxylt.fsf@brailcom.org> Bonjour, >>>>> "N" == Nath writes: [.....] N> do you have an idea of what can cause the problem ? Not yet. Maybe in configuration. Can you please send you speechd.conf? have you nice day -- Jan Buchal Tel: (00420) 24 24 86 008 Mob: (00420) 608023021 _______________________________________________ Speechd mailing list Speechd@lists.freebsoft.org http://lists.freebsoft.org/mailman/listinfo/speechd From hanke@brailcom.org Thu Nov 10 19:03:00 2005 From: Hynek Hanke To: speechd-discuss@nongnu.org Date: Sun Aug 13 12:52:50 2006 Subject: speech-dispatcher + DECtalk on Debian AMD x86-64 : no sound In-Reply-To: <87fyq63mzy.fsf@free.fr> References: <87y840umfx.fsf@free.fr> <86r79rxylt.fsf@brailcom.org> <87fyq63mzy.fsf@free.fr> Message-ID: <1131636044.4007.35.camel@chopin> > But unfortunatly always no sound at all ! Please go to /etc/speech-dispatcher/modules/dtk-generic.conf, find the line # Debug 0 uncomment it and set it to Debug 1 Depending on your version of Speech Dispatcher, you will likely also have to uncomment/modify the line DebugFile and set it to e.g. DebugFile "/tmp/dtk-generic.log" otherwise the logfile will be found in the LogFile directory specified in speechd.conf. Then please restart Speech Dispatcher and try to use the spd-say command two times to say something simple, then send me the resulting logfile to hanke@brailcom.org. This way, we will see what's happening. Thank you, Hynek From nath.ml@free.fr Thu Nov 10 15:38:44 2005 From: Nath To: speechd-discuss@nongnu.org Date: Sun Aug 13 12:52:50 2006 Subject: speech-dispatcher + DECtalk on Debian AMD x86-64 : no sound In-Reply-To: <86ek5pltzo.fsf@brailcom.org> (Jan Buchal's message of "Wed, 09 Nov 2005 16:29:47 +0100") References: <87y840umfx.fsf@free.fr> <86r79rxylt.fsf@brailcom.org> <87fyq63mzy.fsf@free.fr> <86ek5pltzo.fsf@brailcom.org> Message-ID: <87r79obma3.fsf@free.fr> Jan Buchal writes: >>>>>> "[BN" == Nath writes: > > [BN> In fact I modified a little my speechd.conf file because it > [BN> appeared that the modules festival and flite were not commented > [BN> and then speech-dispatcher tried to load them. so now only > [BN> dtk-generic is specified and there is no more errors in the > [BN> /var/log/speech-dispatcher/speech-dispatcher.log file when I > [BN> try to use spd-say. But unfortunatly always no sound at all ! > [BN> For info my soundcard works well since I can hear mp3s without > [BN> trouble and the DECtalk say program works too when I run it > [BN> separatly from speech-dispatcher. As you ask me, here is the > [BN> speechd.conf file (see attachment) > > And speech dispatcher works with flite for example? In any case speech > dispatcher works on amd64 platform, I use it. > > -------------- next part -------------- A non-text attachment was scrubbed... Name: debug-dtk-generic Type: application/octet-stream Size: 3113 bytes Desc: not available Url : http://lists.freebsoft.org/pipermail/speechd/attachments/20051110/6a31ea07/debug-dtk-generic.obj -------------- next part -------------- Yes, on the same machine speech-dispatcher works with flite. But DECtalk doesn't. I attach to this m?l the log file debug-dtk-generic, result from the following commandline : spd-say -o dtk-generic "hello nathalie" Perhaps you will see something wrong in this log file ? Thx, > -- > > Jan Buchal > Tel: (00420) 24 24 86 008 > Mob: (00420) 608023021 > > _______________________________________________ > Speechd mailing list > Speechd@lists.freebsoft.org > http://lists.freebsoft.org/mailman/listinfo/speechd > > -- Nath -------------- next part -------------- _______________________________________________ Speechd mailing list Speechd@lists.freebsoft.org http://lists.freebsoft.org/mailman/listinfo/speechd From buchal@brailcom.org Wed Nov 9 16:29:47 2005 From: Jan Buchal To: speechd-discuss@nongnu.org Date: Sun Aug 13 12:52:50 2006 Subject: speech-dispatcher + DECtalk on Debian AMD x86-64 : no sound In-Reply-To: <87fyq63mzy.fsf@free.fr> (Nath's message of "Tue, 08 Nov 2005 21:25:53 +0100") References: <87y840umfx.fsf@free.fr> <86r79rxylt.fsf@brailcom.org> <87fyq63mzy.fsf@free.fr> Message-ID: <86ek5pltzo.fsf@brailcom.org> >>>>> "[BN" == Nath writes: [BN> In fact I modified a little my speechd.conf file because it [BN> appeared that the modules festival and flite were not commented [BN> and then speech-dispatcher tried to load them. so now only [BN> dtk-generic is specified and there is no more errors in the [BN> /var/log/speech-dispatcher/speech-dispatcher.log file when I [BN> try to use spd-say. But unfortunatly always no sound at all ! [BN> For info my soundcard works well since I can hear mp3s without [BN> trouble and the DECtalk say program works too when I run it [BN> separatly from speech-dispatcher. As you ask me, here is the [BN> speechd.conf file (see attachment) And speech dispatcher works with flite for example? In any case speech dispatcher works on amd64 platform, I use it. -- Jan Buchal Tel: (00420) 24 24 86 008 Mob: (00420) 608023021 _______________________________________________ Speechd mailing list Speechd@lists.freebsoft.org http://lists.freebsoft.org/mailman/listinfo/speechd From hanke@brailcom.org Wed Nov 16 12:27:24 2005 From: Hynek Hanke To: speechd-discuss@nongnu.org Date: Sun Aug 13 12:52:50 2006 Subject: speech-dispatcher + DECtalk on Debian AMD x86-64 : no sound In-Reply-To: <87r79obma3.fsf@free.fr> References: <87y840umfx.fsf@free.fr> <86r79rxylt.fsf@brailcom.org> <87fyq63mzy.fsf@free.fr> <86ek5pltzo.fsf@brailcom.org> <87r79obma3.fsf@free.fr> Message-ID: <1132140423.4123.11.camel@chopin> > But DECtalk doesn't. > I attach to this m?l the log file debug-dtk-generic, result from the > following commandline : spd-say -o dtk-generic "hello nathalie" Hello Nath, from the logfile it looks like everything works correctly, but the Dectalks say command doesn't do anything and returns immediatelly with a strange error code of 32512. Could you please try to run these two commands: echo "[:np][:ra 338][:dv ap 225]" >/tmp/dtk-speak.txt && echo "hello nathalie" | fmt >>/tmp/dtk-speak.txt say -fi /tmp/dtk-speak.txt Does it work? If not, what's in /tmp/dtk-speak.txt. If yes, does it also work under the speech-dispatcher user? With regards, Hynek Hanke From hanke@brailcom.org Thu Nov 10 19:03:00 2005 From: Hynek Hanke To: speechd-discuss@nongnu.org Date: Sun Aug 13 12:52:50 2006 Subject: speech-dispatcher + DECtalk on Debian AMD x86-64 : no sound In-Reply-To: <87fyq63mzy.fsf@free.fr> References: <87y840umfx.fsf@free.fr> <86r79rxylt.fsf@brailcom.org> <87fyq63mzy.fsf@free.fr> Message-ID: <1131636044.4007.35.camel@chopin> > But unfortunatly always no sound at all ! Please go to /etc/speech-dispatcher/modules/dtk-generic.conf, find the line # Debug 0 uncomment it and set it to Debug 1 Depending on your version of Speech Dispatcher, you will likely also have to uncomment/modify the line DebugFile and set it to e.g. DebugFile "/tmp/dtk-generic.log" otherwise the logfile will be found in the LogFile directory specified in speechd.conf. Then please restart Speech Dispatcher and try to use the spd-say command two times to say something simple, then send me the resulting logfile to hanke@brailcom.org. This way, we will see what's happening. Thank you, Hynek From hanke@brailcom.org Wed Dec 7 10:03:41 2005 From: Hynek Hanke To: speechd-discuss@nongnu.org Date: Sun Aug 13 12:52:51 2006 Subject: Speech-dispatcher initialization In-Reply-To: <20051207.091350.102550073.pollock.nageoire@wanadoo.fr> References: <20051207.091350.102550073.pollock.nageoire@wanadoo.fr> Message-ID: <1133946192.4033.8.camel@chopin> Pierre Lorenzon p??e v St 07. 12. 2005 v 09:13 +0100: > I sent this message two weeks ago but got no > answer. Is it so strange that there no answer ? > Do you want more details about the machine architecture, > the gcc version or whatever ? Should I try to > understand where the error comes from ? Pierre, I'm sorry for the delay. I examined the logs after you sent them and I don't see how this can happen with the current code. Since we were planning to release the 0.6 version of Speech Dispatcher right now, I thought to send you the 0.6-rc1 tarball as soon as it is ready so that you can test it and see if this issue still happens. However, there were some unexpected delays and 0.6-rc1 is still not public. I could either send you a tarball with the latest code privately to your email or you could try to use the latest CVS version. The first has the advantage that you don't have to build it with automake, autoconf etc., you just compile it. Please let me know what do you prefer. With Regards, Hynek From nath.ml@free.fr Mon Nov 7 23:23:30 2005 From: Nath To: speechd-discuss@nongnu.org Date: Sun Aug 13 12:52:51 2006 Subject: speech-dispatcher + DECtalk on Debian AMD x86-64 : no sound Message-ID: <87y840umfx.fsf@free.fr> Hi all, I try to have speech-dispatcher working on my debian box (testing release on AMD 64 bit architecture) with DECtalk software but have no luck. all seems to be installed well, speech-dispatcher is started and the additional module sd_generic is loaded. I installed speech-dispatcher using Debian packages. so the speech-dispatcher version installed is 0.5-3. I also installed libspeechd1. I tried with DECtalk 4.61 and 5.0 but no sound at all. To test, I try to use spd-say and run the following commandline : spd-say -o dtk-generic "hello" or spd-say "hello" No error are returned. The file /tmp/dtk-speak.txt seems to be correct : -------------- [:np][:ra 338][:dv ap 225] hello ------------------------ But something seems to be wrong in the speech-dispatcher.log file : -------------------- [Mon Nov 7 22:11:55 2005 : 91524] speechd: Warning: Didn't find prefered output module, using default [Mon Nov 7 22:11:55 2005 : 91549] speechd: Error: Can't find default output module or it's not working! [Mon Nov 7 22:11:55 2005 : 91554] speechd: Error: Unspecified output module! [Mon Nov 7 22:11:55 2005 : 91559] speechd: Output module doesn't work... [Mon Nov 7 22:11:55 2005 : 91564] speechd: Error: Output module failed [Mon Nov 7 22:11:55 2005 : 91569] speechd: Output module working status: 0 (pid:1) [Mon Nov 7 22:11:55 2005 : 91575] speechd: Output module terminated abnormally, probably crashed. [Mon Nov 7 22:11:55 2005 : 91623] speechd: Connection closed ------------------------- In speechd.conf I commented all the AddModule lines except for the dtk-generic one. a precision : since It doesn't exist a 64 bit DECtalk software version I use the say command from the 32 bits version and the command works itselve on the commandline since I have installed the ia32-libs package on my system. do you have an idea of what can cause the problem ? thx in advance for your help ! -- Nath _______________________________________________ Speechd mailing list Speechd@lists.freebsoft.org http://lists.freebsoft.org/mailman/listinfo/speechd From pollock.nageoire@wanadoo.fr Wed Dec 7 11:47:28 2005 From: Pierre Lorenzon To: speechd-discuss@nongnu.org Date: Sun Aug 13 12:52:51 2006 Subject: Speech-dispatcher initialization In-Reply-To: <1133946192.4033.8.camel@chopin> References: <20051207.091350.102550073.pollock.nageoire@wanadoo.fr> <1133946192.4033.8.camel@chopin> Message-ID: <20051207.114728.35687383.pollock.nageoire@wanadoo.fr> Hi, Thank you for your answer. I am not so surprised that this error is unexpected since we tried to run speech-dispatcher on an 64bit architecture and I am not convinced that all the tools works fine. I think that we will try to use the cvs version ; since automake autoconf and succh tools are installed and I am used to use these tools. Best regards Pierre From: Hynek Hanke To: speechd-discuss@nongnu.org Subject: Re: Speech-dispatcher initialization Date: Wed, 07 Dec 2005 10:03:12 +0100 hanke> Pierre Lorenzon p??e v St 07. 12. 2005 v 09:13 +0100: hanke> > I sent this message two weeks ago but got no hanke> > answer. Is it so strange that there no answer ? hanke> > Do you want more details about the machine architecture, hanke> > the gcc version or whatever ? Should I try to hanke> > understand where the error comes from ? hanke> hanke> Pierre, hanke> hanke> I'm sorry for the delay. I examined the logs after you sent them and I hanke> don't see how this can happen with the current code. Since we were hanke> planning to release the 0.6 version of Speech Dispatcher right now, hanke> I thought to send you the 0.6-rc1 tarball as soon as it is ready hanke> so that you can test it and see if this issue still happens. However, hanke> there were some unexpected delays and 0.6-rc1 is still not public. hanke> hanke> I could either send you a tarball with the latest code privately to your hanke> email or you could try to use the latest CVS version. The first has the hanke> advantage that you don't have to build it with automake, autoconf etc., hanke> you just compile it. Please let me know what do you prefer. hanke> hanke> With Regards, hanke> Hynek hanke> hanke> B _______________________________________________ Speechd mailing list Speechd@lists.freebsoft.org http://lists.freebsoft.org/mailman/listinfo/speechd From hanke@brailcom.org Thu Mar 30 23:26:20 2006 From: Hynek Hanke To: speechd-discuss@nongnu.org Date: Sun Aug 13 12:52:49 2006 Subject: Documentation In-Reply-To: <200603292127.03239.garycramblitt@comcast.net> References: <200603292127.03239.garycramblitt@comcast.net> Message-ID: <1143753979.4238.38.camel@chopin> Gary Cramblitt p??e v St 29. 03. 2006 v 21:27 -0500: > I have committed spelling and grammar corrections for speech-dispatcher.texi. Thank you. > When I modify the html_docs line in doc/Makefile.am from > html_docs = speechd.html ssip.html spd-say.texi > to > html_docs = speech-dispatcher.html ssip.html spd-say.texi > then I do not get the error. This fix is correct. Please commit it. > 2. When generating a pdf of the manual, the figures/architecture.pdf file is > too wide for the page. This causes a blank area at the point where the file > is referenced and part of the image appears a page or two later. I committed > a smaller version of architecture.pdf to fix this problem, but it is quite > fuzzy. I'd be grateful if someone could figure out how to scale the image > down but retain sharpness. It might be better to generate a smaller postscript version from the .dia source with Dia and then convert it to PDF. Thanks again, Hynek From pollock.nageoire@wanadoo.fr Fri Feb 24 14:40:19 2006 From: Pierre Lorenzon To: speechd-discuss@nongnu.org Date: Sun Aug 13 12:52:49 2006 Subject: prblem with w3 In-Reply-To: References: Message-ID: <20060224.144019.71100772.pollock.nageoire@wanadoo.fr> Hello, I would personnaly suggest to use w3m instead of w3 and its emacs extension emacs-w3m see : http://emacs-w3m.namazu.org Bests Pierre _______________________________________________ Speechd mailing list Speechd@lists.freebsoft.org http://lists.freebsoft.org/mailman/listinfo/speechd From mooncar@libero.it Fri Feb 17 11:29:35 2006 From: mooncar@libero.it To: speechd-discuss@nongnu.org Date: Sun Aug 13 12:52:49 2006 Subject: speechd problem Message-ID: ok thank you! I'm going to try with cvs. Best regards veronica. > >>>>> "mi" == mooncar\@libero\ it writes: > > mi> SIOD ERROR: wrong type of argument to times SIOD ERROR: bad > mi> function : "Tree_ZScores" > > festival-freebsoft-utils 0.6 doesn't work with Italian. This has been > fixed in festival-freebsoft-utils CVS, see > http://www.freebsoft.org/festival-freebsoft-utils for instructions how > to access festival-freebsoft-utils CVS. > > Regards, > > Milan Zamazal > > _______________________________________________ Speechd mailing list Speechd@lists.freebsoft.org http://lists.freebsoft.org/mailman/listinfo/speechd From tcross@rapttech.com.au Wed Mar 15 09:05:50 2006 From: Tim Cross To: speechd-discuss@nongnu.org Date: Sun Aug 13 12:52:49 2006 Subject: shell under emacs In-Reply-To: References: Message-ID: <17431.51934.437081.475352@tiger.rapttech.com.au> Eeeek! What was I thinking. The shell command history is reached using M-p and M-n not C-p and C-n. You may also be able to use C- and C-. Just got home from a bad day at work and wasn't typing what I was thinking! Tim mooncar@libero.it writes: > > Hi! > I have installed emacs with speech support,now it works well,but the boy > who is using that tells me he needs to use the command M-x shell, in emacs. > Is it possible to recall the commands that he have just issued, > like in the linux shell out of emacs?now he must re-type the commands... > I have tryied to find a solution but I don't use that program-I only > installed and tryied if it's working- so if someone can help... > thanks in advance, > veronica > > _______________________________________________ > Speechd mailing list > Speechd@lists.freebsoft.org > http://lists.freebsoft.org/mailman/listinfo/speechd _______________________________________________ Speechd mailing list Speechd@lists.freebsoft.org http://lists.freebsoft.org/mailman/listinfo/speechd From pollock.nageoire@wanadoo.fr Mon Feb 27 22:35:32 2006 From: Pierre Lorenzon To: speechd-discuss@nongnu.org Date: Sun Aug 13 12:52:49 2006 Subject: problem with w3 and w3m In-Reply-To: References: Message-ID: <20060227.223532.74743889.pollock.nageoire@wanadoo.fr> Hi Veronica, Yes you are missing gc ; which is required by w3m. http://www.hpl.hp.com/personal/Hans_Boehm/gc Bests Pierre _______________________________________________ Speechd mailing list Speechd@lists.freebsoft.org http://lists.freebsoft.org/mailman/listinfo/speechd From hanke@brailcom.org Thu Mar 30 23:26:18 2006 From: Hynek Hanke To: speechd-discuss@nongnu.org Date: Sun Aug 13 12:52:49 2006 Subject: Documentation In-Reply-To: <200603292127.03239.garycramblitt@comcast.net> References: <200603292127.03239.garycramblitt@comcast.net> Message-ID: <1143753979.4238.38.camel@chopin> Gary Cramblitt p??e v St 29. 03. 2006 v 21:27 -0500: > I have committed spelling and grammar corrections for speech-dispatcher.texi. Thank you. > When I modify the html_docs line in doc/Makefile.am from > html_docs = speechd.html ssip.html spd-say.texi > to > html_docs = speech-dispatcher.html ssip.html spd-say.texi > then I do not get the error. This fix is correct. Please commit it. > 2. When generating a pdf of the manual, the figures/architecture.pdf file is > too wide for the page. This causes a blank area at the point where the file > is referenced and part of the image appears a page or two later. I committed > a smaller version of architecture.pdf to fix this problem, but it is quite > fuzzy. I'd be grateful if someone could figure out how to scale the image > down but retain sharpness. It might be better to generate a smaller postscript version from the .dia source with Dia and then convert it to PDF. Thanks again, Hynek _______________________________________________ Speechd mailing list Speechd@lists.freebsoft.org http://lists.freebsoft.org/mailman/listinfo/speechd From buchal@brailcom.org Tue Jan 24 08:17:26 2006 From: Jan Buchal To: speechd-discuss@nongnu.org Date: Sun Aug 13 12:52:49 2006 Subject: Need help with festival in red hat In-Reply-To: (mooncar@libero.it's message of "Mon, 23 Jan 2006 21:25:03 +0100") References: Message-ID: <86vewaksc9.fsf@brailcom.org> >>>>> "mi" == mooncar\@libero\ it writes: mi> Hello! I have to install speechd with festival on a computer mi> with Linux Red Hat 8. I have already installed FESTIVAL by mi> tarball, but it speak too fast. The exact procedure that I have mi> followed is [....] mi> It seems that all goes well, but when I run Festival it speak mi> too fast, and if I use the command (Parameter.set mi> 'Duration_Stretch 2) the result is that pronounciation in mi> stretched, but still deformed.. Why? Has someone any idea ? Do you mean too fast only or with speed is increase pitch too? Which sound system do you use? OSS or ALSA? Which sound card? How works flite? -- Jan Buchal Tel: (00420) 24 24 86 008 Mob: (00420) 608023021 _______________________________________________ Speechd mailing list Speechd@lists.freebsoft.org http://lists.freebsoft.org/mailman/listinfo/speechd From mooncar@libero.it Fri Feb 17 11:08:09 2006 From: mooncar@libero.it To: speechd-discuss@nongnu.org Date: Sun Aug 13 12:52:49 2006 Subject: speechd problem Message-ID: > mooncar@libero.it p?se v Ct 16. 02. 2006 v 18:34 +0100: > > How can I see wich version I am using? > > I don't know about any general way how to determine this. > aplay --version aplay: version 1.0.6 by Jaroslav Kysela > > I have file > > libasound.so.2.0.0 and i have scientific linux c 4 > > I'm using the same version and it works for me. The bug is at least > partially in ALSA. The library is not supposed to halt the entire > process on assertion. I'll have to talk about this with ALSA developers, > so I fear there is no immediate help for you. > > You might still use OSS for the sound output. See festival conf under > etc/speech-dispatcher/modules/. ok with OSS it works, at least in english...thank you! > > Can you please tell me what soundcard you use? What > architecture is your system running? Processor: cat /proc/cpuinfo processor : 0 vendor_id : AuthenticAMD cpu family : 6 model : 8 model name : AMD Athlon(tm) XP 2600+ stepping : 1 cpu MHz : 2144.873 cache size : 256 KB fdiv_bug : no hlt_bug : no f00f_bug : no coma_bug : no fpu : yes fpu_exception : yes cpuid level : 1 wp : yes flags : fpu vme de pse tsc msr pae mce cx8 apic mtrr pge mca cmov pat pse36 mmx fxsr sse syscall mmxext 3dnowext 3dnow bogomips : 4227.07 and sound card: VIA Technologies, VT8233 , ac 97 audio controller But there is another problem... now, speech dispatcher is ok I installed speechd-el by copying *.el files in the emacs load path , and it's ok- in english. then I want to change to italian language. I add (it italian) on voice-select.scm, I add LanguageDefaultModule "it" "festival" on speechd.conf, but when I change to italian language, emacs became silent and I have SIOD ERROR: wrong type of argument to times SIOD ERROR: bad function : "Tree_ZScores" What is wrong? Best regards, veronica > > With regards, > Hynek Hanke > > _______________________________________________ Speechd mailing list Speechd@lists.freebsoft.org http://lists.freebsoft.org/mailman/listinfo/speechd From pdm@freebsoft.org Fri Feb 17 11:15:08 2006 From: Milan Zamazal To: speechd-discuss@nongnu.org Date: Sun Aug 13 12:52:49 2006 Subject: speechd problem In-Reply-To: (mooncar@libero.it's message of "Fri, 17 Feb 2006 11:08:09 +0100") References: Message-ID: <87pslms34j.fsf@blackbird.zamazal.org> >>>>> "mi" == mooncar\@libero\ it writes: mi> SIOD ERROR: wrong type of argument to times SIOD ERROR: bad mi> function : "Tree_ZScores" festival-freebsoft-utils 0.6 doesn't work with Italian. This has been fixed in festival-freebsoft-utils CVS, see http://www.freebsoft.org/festival-freebsoft-utils for instructions how to access festival-freebsoft-utils CVS. Regards, Milan Zamazal _______________________________________________ Speechd mailing list Speechd@lists.freebsoft.org http://lists.freebsoft.org/mailman/listinfo/speechd From hanke@brailcom.org Thu Feb 16 17:34:08 2006 From: Hynek Hanke To: speechd-discuss@nongnu.org Date: Sun Aug 13 12:52:49 2006 Subject: error while loading shared libraries: libdotconf In-Reply-To: <86zmkr9twn.fsf@brailcom.org> References: <86zmkr9twn.fsf@brailcom.org> Message-ID: <1140107645.4253.42.camel@chopin> > mi> I have just installed speech dispatcher 0.6 from tarball, the > mi> installation was successfully but when I run speech-dispatcher I > mi> have this error > mi> speech-dispatcher: error while loading shared libraries: > mi> libdotconf-1.0.so.0: cannot open shared object file: No such > mi> file or directory > Try install speechd again but before remove all libspeechd. I'm not sure > if it help.... No, I don't think this will help. It has nothing to do with libspeechd and removing all libspeechd is not a good idea because some applications on the system might be linked against older versions of libspeechd and this would break them. mooncar, I think there is some mess with the instalation of libdotconf. The configure script for Dispatcher performs a check if the library is available and from your writing it seems it passed the check. You also say you have libdotconf installed. Maybe you did configure as another user as the one who is trying to do make or something like that? I'd guess that gcc is not looking into /usr/local/lib for the library and only looks into /usr/lib. Try compiling again, starting with make clean LDFLAGS="-L/usr/local/lib" ./configure If it doesn't help, do make clean ./configure 2&> configure.log make 2&> make.log and send me the two .log files so that I can see what is wrong. Have a nice day, Hynek From garycramblitt@comcast.net Fri Mar 31 01:45:20 2006 From: Gary Cramblitt To: speechd-discuss@nongnu.org Date: Sun Aug 13 12:52:50 2006 Subject: Documentation In-Reply-To: <1143753979.4238.38.camel@chopin> References: <200603292127.03239.garycramblitt@comcast.net> <1143753979.4238.38.camel@chopin> Message-ID: <200603301845.20522.garycramblitt@comcast.net> On Thursday 30 March 2006 16:26, Hynek Hanke wrote: > > When I modify the html_docs line in doc/Makefile.am from > > html_docs = speechd.html ssip.html spd-say.texi > > to > > html_docs = speech-dispatcher.html ssip.html spd-say.texi > > then I do not get the error. > > This fix is correct. Please commit it. Will do. Thanks. > > 2. When generating a pdf of the manual, the figures/architecture.pdf > > file is too wide for the page. This causes a blank area at the point > > where the file is referenced and part of the image appears a page or two > > later. I committed a smaller version of architecture.pdf to fix this > > problem, but it is quite fuzzy. I'd be grateful if someone could figure > > out how to scale the image down but retain sharpness. > > It might be better to generate a smaller postscript version from > the .dia source with Dia and then convert it to PDF. I tried that. dia doesn't have an export plugin to pdf format. When I export to Encapsulated Postscript (.eps), it doesn't offer an option to size the output, so I end up with essentially the existing architecture.eps already in the repository, and when I resize that with ImageMagick, it gets fuzzy. I also tried exporting to png, which does offer an output size, but after conversion to pdf I get a "page" with the image in the lower left corner. I figured maybe I could size the figure in dia, then export, but I could not find any way to size the diagram in dia. I also tried export to svg, but when I load the resulting file into karbon or konqueror, the fonts are just black squares. I also tried to use scribus with no luck. Sorry to be a pain. The image I committed isn't bad, but it can be better. :/ I'm hoping some good samaritan will take a shot at it. There's gotta be a way. Afterall dia is a vector based image, so it should be possible to get a crystal-clear image. Needs to be pdf of a image about 450x155 pixels if anyone wants to give it a go. -- Gary Cramblitt (aka PhantomsDad) _______________________________________________ Speechd mailing list Speechd@lists.freebsoft.org http://lists.freebsoft.org/mailman/listinfo/speechd From hanke@brailcom.org Thu Feb 16 18:26:41 2006 From: Hynek Hanke To: speechd-discuss@nongnu.org Date: Sun Aug 13 12:52:50 2006 Subject: speechd problem In-Reply-To: References: Message-ID: <1140110798.4253.51.camel@chopin> mooncar@libero.it p??e v ?t 16. 02. 2006 v 18:00 +0100: > the festival.log file is > [...] > Thu Feb 16 17:45:02 2006 [786326] ALSA: Filling with silence up to the period size, silent_samples=1 > pcm.c:5958: snd_pcm_mmap_commit: Assertion `frames <= snd_pcm_mmap_avail(pcm)' failed. Hm, it looks like some problem comming from inside ALSA which stops Dispatchers output module. Which version of libasound are you using? With regards, Hynek From pdm@freebsoft.org Wed Mar 15 10:04:21 2006 From: Milan Zamazal To: speechd-discuss@nongnu.org Date: Sun Aug 13 12:52:50 2006 Subject: shell under emacs In-Reply-To: <17431.51662.617759.229441@tiger.rapttech.com.au> (Tim Cross's message of "Wed, 15 Mar 2006 19:01:18 +1100") References: <17431.51662.617759.229441@tiger.rapttech.com.au> Message-ID: <8764mg2i5m.fsf@blackbird.zamazal.org> >>>>> "TC" == Tim Cross writes: TC> However, note that you probably better off using M-x term rather TC> than M-x shell. To clarify this: `M-x term' and `M-x shell' are two different things. `M-x term' runs an (imperfect) terminal emulator, while `M-x shell' runs a command line interface to shells. If you want to run a visual program or a program with its own command completion, you need to run `M-x term' which redirects all the user interaction to the program. If you need to run commands from shell, `M-x shell' allows you, unlike `M-x term', to use all the standard Emacs commands to edit command lines, read command outputs, search in them, handle command history, etc. As Tim has mentioned, if you want to manage files, it's useful to learn Dired and Tramp. Regards, Milan Zamazal ----------------------------------------------------------------------------- To unsubscribe from the emacspeak list or change your address on the emacspeak list send mail to "emacspeak-request@cs.vassar.edu" with a subject of "unsubscribe" or "help" From mooncar@libero.it Thu Feb 16 18:34:26 2006 From: mooncar@libero.it To: speechd-discuss@nongnu.org Date: Sun Aug 13 12:52:50 2006 Subject: speechd problem Message-ID: How can I see wich version I am using? I have file libasound.so.2.0.0 and i have scientific linux c 4 > mooncar@libero.it p?se v Ct 16. 02. 2006 v 18:00 +0100: > > the festival.log file is > > [...] > > Thu Feb 16 17:45:02 2006 [786326] ALSA: Filling with silence up to the period size, silent_samples=1 > > pcm.c:5958: snd_pcm_mmap_commit: Assertion `frames <= snd_pcm_mmap_avail(pcm)' failed. > > Hm, it looks like some problem comming from inside ALSA > which stops Dispatchers output module. Which version of > libasound are you using? > > With regards, > Hynek > > _______________________________________________ Speechd mailing list Speechd@lists.freebsoft.org http://lists.freebsoft.org/mailman/listinfo/speechd From mooncar@libero.it Mon Feb 27 17:02:23 2006 From: mooncar@libero.it To: speechd-discuss@nongnu.org Date: Sun Aug 13 12:52:50 2006 Subject: problem with w3 and w3m Message-ID: Done, but during the ./configure script I have this checking GC library exists... yes checking gc.h usability... no checking gc.h presence... no checking for gc.h... no checking GC header location... /usr /usr/local /root checking /usr/include... checking gc.h usability... no [...] checking for gc.h... no configure: error: gc.h not found Maybe I've missing something? > > Hello, > > I would personnaly suggest to use w3m instead of w3 and > its emacs extension emacs-w3m see : > > http://emacs-w3m.namazu.org > > Bests > > Pierre > > > _______________________________________________ Speechd mailing list Speechd@lists.freebsoft.org http://lists.freebsoft.org/mailman/listinfo/speechd From mooncar@libero.it Thu Feb 23 16:11:36 2006 From: mooncar@libero.it To: speechd-discuss@nongnu.org Date: Sun Aug 13 12:52:50 2006 Subject: prblem with w3 Message-ID: Hello, I've a problem with w3... I have installed w3 from tarball. If I run ./configure without options, the Makefile is [....] prefix = /usr datadir = ${prefix}/share lispdir = $(prefix)/share/xemacs/site-lisp infodir = ${prefix}/info [...] and w3 is installed in /usr/share/xemacs/site-lisp, then I can use it from xemacs, but not from emacs. If i run ./configure --lispdir=/usr/local/share/emacs/site-lisp, when I type M-x load-library RET w3, I have: file /usr/local/share/emacs/site-lisp/w3.elc was not compiled in emacs. How can I resolve?? thank you! _______________________________________________ Speechd mailing list Speechd@lists.freebsoft.org http://lists.freebsoft.org/mailman/listinfo/speechd From mooncar@libero.it Thu Feb 23 14:36:40 2006 From: mooncar@libero.it To: speechd-discuss@nongnu.org Date: Sun Aug 13 12:52:50 2006 Subject: question about speech dispatcher Message-ID: Hi to all! I have on my pc speech dispatcher, with festival and speechd-el for emacs. It works well, and I have italian language too. But I have some questions: 1)if I turn on my pc, I run festival --server, then speech-dispatcher, then emacs, it's ok, because festival server accept the connection from speech dispatcher. If I log off, and then I log on, but I don't restart linux, when I start Speech-dispatcher I have: >Speech-dispatcher already running so speech dispatcher don't communicate with festival --server (there is non connection between them). Must I restart my pc every time? Is there one way for restart speech dispatcher, without restart linux? 2) I must start speech-dispatcher from root. Can I use speech dispatcher from simple user? I don't want tell the root password whoever... thank you! _______________________________________________ Speechd mailing list Speechd@lists.freebsoft.org http://lists.freebsoft.org/mailman/listinfo/speechd From buchal@brailcom.org Tue Jan 24 12:28:41 2006 From: buchal@brailcom.org To: speechd-discuss@nongnu.org Date: Sun Aug 13 12:52:50 2006 Subject: Need help with festival in red hat In-Reply-To: (mooncar@libero.it's message of "Tue, 24 Jan 2006 11:19:02 +0100") References: Message-ID: <864q3tdfva.fsf@brailcom.org> >>>>> "mi" == mooncar\@libero\ it writes: mi> With speed is increase pitch too. >> Which sound system do you use? OSS or ALSA? mi> I think OSS but I'm not sure..How can I check it? lsmod as root. If you use oss, try alsa. There are modules snd- >> Which sound card? mi> VT8233 AC97 Audio Controller, by VIA Technologies OK, it seems that the problem is with module which support this sound card not well. >> How works flite? mi> It speaks too fast, like festival; but the sound is also very mi> noisy, so I don't know if the pitch is increased too-I don't mi> understand hardly anything of what it speaks... Yes, yes I think will be so. So try the alsa system and we will see. p.s. This email come on my personal email only. I'm not sure if you wanted it. -- Jan Buchal Tel: (00420) 24 24 86 008 Mob: (00420) 608023021 http://www.brailcom.org http://www.nepodpis.cz http://internet.proslepe.cz http://www.freebsoft.org http://wahelper.brailcom.org _______________________________________________ Speechd mailing list Speechd@lists.freebsoft.org http://lists.freebsoft.org/mailman/listinfo/speechd From mooncar@libero.it Tue Feb 28 12:05:17 2006 From: mooncar@libero.it To: speechd-discuss@nongnu.org Date: Sun Aug 13 12:52:50 2006 Subject: resolved w3m Message-ID: resolved by installing it by rpm! I have already gc but the problem was header files! thank you again, regards veronica > > Hi Veronica, > > Yes you are missing gc ; which is required by > w3m. > > http://www.hpl.hp.com/personal/Hans_Boehm/gc > > > Bests > > Pierre > > _______________________________________________ Speechd mailing list Speechd@lists.freebsoft.org http://lists.freebsoft.org/mailman/listinfo/speechd From pollock.nageoire@wanadoo.fr Thu Feb 23 15:11:57 2006 From: Pierre Lorenzon To: speechd-discuss@nongnu.org Date: Sun Aug 13 12:52:50 2006 Subject: question about speech dispatcher In-Reply-To: References: Message-ID: <20060223.151157.41644357.pollock.nageoire@wanadoo.fr> Hi, Indeed, you don't need to start speech-dispatcher each time you log in. Festival server and speech-dispatcher can run in the background during all the time your pc is running just like other server applications like apache server or whatever. Then each time you start emacs with speechd-el client, it connects to the running speech-dispatcher. I don't know for what you're using your pc but if there is no particular application which need alot of resources, you can let festival and speech-dispatcher run in the background. In this case, the best way for me is to start them in the startup scripts (in /etc/rc.d/rc3.d ... in most of the distributions.) Best regards Pierre _______________________________________________ Speechd mailing list Speechd@lists.freebsoft.org http://lists.freebsoft.org/mailman/listinfo/speechd From mooncar@libero.it Thu Feb 16 18:00:28 2006 From: mooncar@libero.it To: speechd-discuss@nongnu.org Date: Sun Aug 13 12:52:50 2006 Subject: speechd problem Message-ID: Resolved the problem with dotconf library, I have added the path ed executed ldconfig, but I have another problem I run festival --server, then speech-dispatcher, I have server Thu Feb 16 17:44:33 2006 : Festival server started on port 1314 client(1) Thu Feb 16 17:44:39 2006 : accepted from localhost.localdomain then I type spd-say "hello" and i have client(1) Thu Feb 16 17:45:02 2006 : disconnected the festival.log file is [...] Thu Feb 16 17:45:02 2006 [786326] ALSA: Filling with silence up to the period size, silent_samples=1 pcm.c:5958: snd_pcm_mmap_commit: Assertion `frames <= snd_pcm_mmap_avail(pcm)' failed. and the speechd.log: [...] Communicating with Festival through a socket. --------------- [Thu Feb 16 17:44:39 2006 : 527646] speechd: Reading configuration for pattern emacs:* [Thu Feb 16 17:44:39 2006 : 527950] speechd: Reading configuration for pattern *:gnomespeech:* [Thu Feb 16 17:44:39 2006 : 527993] speechd: Configuration has been read from "/usr/local/etc/speech-dispatcher//speechd.conf" [Thu Feb 16 17:44:39 2006 : 528011] speechd: Speech Dispatcher started with 1 output module [Thu Feb 16 17:44:39 2006 : 528064] speechd: Openning socket connection [Thu Feb 16 17:44:39 2006 : 528894] speechd: Speech Dispatcher waiting for clients ... [Thu Feb 16 17:45:02 2006 : 3879] speechd: Connection closed help... _______________________________________________ Speechd mailing list Speechd@lists.freebsoft.org http://lists.freebsoft.org/mailman/listinfo/speechd From hanke@brailcom.org Thu Feb 23 19:52:32 2006 From: Hynek Hanke To: speechd-discuss@nongnu.org Date: Sun Aug 13 12:52:50 2006 Subject: question about speech dispatcher In-Reply-To: References: Message-ID: <1140720753.4197.11.camel@chopin> >Must I restart my pc every time? Is there one way for restart speech dispatcher, >without restart linux? Hi Veronica, sure there is a way. You just kill speech-dispatcher (or send it a SIGTERM) signal and then run it again. Like this killall speech-dispatcher speech-dispatcher When you start speech-dispatcher, it runs in the background, so you can't start a new one (it complains that one is already running). > 2) I must start speech-dispatcher from root. Can I use speech dispatcher > from simple user? I don't want tell the root password whoever... Yes, you can do that. But I think a better approach would be to start Festival and Speech Dispatcher automatically as system services so that you don't have to care about this at all. The exact steps to do this are distribution specific. Have a nice day, Hynek Hanke _______________________________________________ Speechd mailing list Speechd@lists.freebsoft.org http://lists.freebsoft.org/mailman/listinfo/speechd From mooncar@libero.it Tue Jan 24 11:19:02 2006 From: mooncar@libero.it To: speechd-discuss@nongnu.org Date: Sun Aug 13 12:52:50 2006 Subject: Need help with festival in red hat Message-ID: > >>>>> "mi" == mooncar\@libero\ it writes: > > mi> Hello! I have to install speechd with festival on a computer > mi> with Linux Red Hat 8. I have already installed FESTIVAL by > mi> tarball, but it speak too fast. The exact procedure that I have > mi> followed is > [....] > > > mi> It seems that all goes well, but when I run Festival it speak > mi> too fast, and if I use the command (Parameter.set > mi> 'Duration_Stretch 2) the result is that pronounciation in > mi> stretched, but still deformed.. Why? Has someone any idea ? > Do you mean too fast only or with speed is increase pitch too? With speed is increase pitch too. > Which sound system do you use? OSS or ALSA? I think OSS but I'm not sure..How can I check it? >Which sound card? VT8233 AC97 Audio Controller, by VIA Technologies >How works flite? It speaks too fast, like festival; but the sound is also very noisy, so I don't know if the pitch is increased too-I don't understand hardly anything of what it speaks... > > > -- > > Jan Buchal > Tel: (00420) 24 24 86 008 > Mob: (00420) 608023021 > > _______________________________________________ Speechd mailing list Speechd@lists.freebsoft.org http://lists.freebsoft.org/mailman/listinfo/speechd From garycramblitt@comcast.net Thu Feb 16 22:42:08 2006 From: Gary Cramblitt To: speechd-discuss@nongnu.org Date: Sun Aug 13 12:52:50 2006 Subject: speechd problem In-Reply-To: References: Message-ID: <200602161642.08697.garycramblitt@comcast.net> On Thursday 16 February 2006 12:34, mooncar@libero.it wrote: > How can I see wich version I am using? I have file libasound.so.2.0.0 and i > have scientific linux c 4 aplay --version or locate alsa/version.h and examine file for string #define SND_LIB_VERSION_STR -- Gary Cramblitt (aka PhantomsDad) _______________________________________________ Speechd mailing list Speechd@lists.freebsoft.org http://lists.freebsoft.org/mailman/listinfo/speechd From mooncar@libero.it Thu Feb 16 15:41:04 2006 From: mooncar@libero.it To: speechd-discuss@nongnu.org Date: Sun Aug 13 12:52:50 2006 Subject: error while loading shared libraries: libdotconf Message-ID: Hello! I have just installed speech dispatcher 0.6 from tarball, the installation was successfully but when I run speech-dispatcher I have this error speech-dispatcher: error while loading shared libraries: libdotconf-1.0.so.0: cannot open shared object file: No such file or directory but I've already installed dotconf 1.0.13, and the file libdotconf-1.0.so.0 is in /usr/local/lib. What can I do so that speech-dispatcher can find him? Thanks _______________________________________________ Speechd mailing list Speechd@lists.freebsoft.org http://lists.freebsoft.org/mailman/listinfo/speechd From buchal@brailcom.org Thu Feb 16 16:59:20 2006 From: Jan Buchal To: speechd-discuss@nongnu.org Date: Sun Aug 13 12:52:51 2006 Subject: error while loading shared libraries: libdotconf In-Reply-To: (mooncar@libero.it's message of "Thu, 16 Feb 2006 15:41:04 +0100") References: Message-ID: <86zmkr9twn.fsf@brailcom.org> >>>>> "mi" == mooncar\@libero\ it writes: mi> Hello! mi> I have just installed speech dispatcher 0.6 from tarball, the mi> installation was successfully but when I run speech-dispatcher I mi> have this error mi> speech-dispatcher: error while loading shared libraries: mi> libdotconf-1.0.so.0: cannot open shared object file: No such mi> file or directory Try install speechd again but before remove all libspeechd. I'm not sure if it help.... -- Jan Buchal Tel: (00420) 24 24 86 008 Mob: (00420) 608023021 _______________________________________________ Speechd mailing list Speechd@lists.freebsoft.org http://lists.freebsoft.org/mailman/listinfo/speechd From garycramblitt@comcast.net Fri Mar 31 23:44:58 2006 From: Gary Cramblitt To: speechd-discuss@nongnu.org Date: Sun Aug 13 12:52:51 2006 Subject: Documentation In-Reply-To: <442CE612.8020503@brailcom.org> References: <200603292127.03239.garycramblitt@comcast.net> <442CC00F.6030207@brailcom.org> <442CE612.8020503@brailcom.org> Message-ID: <200603311644.58103.garycramblitt@comcast.net> On Friday 31 March 2006 03:19, Tomas Cerha wrote: > Tomas Cerha wrote: > > Hi, I produced all the previous formats from the original dia drawing, > > so I believe I'll be able to do that again. I don't remember how I did > > it, but I'll give it a try... > > Hello, > > the solution was quite simple at the end. I simply converted the EPS to > a vector graphics PDF using epstopdf and added an explicit image width > 155mm to the texi source file. The image is automatically scaled by TeX > to fit the given width and the result remains a vector graphics, so it > will retain the full resolution when printed. I looked for an eps2pdf program but could only find them for Windows and I'm running Debian Linux. Could you convert the file for me and send it? It might work better for me. I can use ImageMagick to convert an eps file to pdf, but when I add 155mm to the texi source as you suggest, it is ignored. The result is only part of the image fits on the page and texinfo mangles the document pretty badly. According to my texinfo manual, Version 3.141592-1.21a-2.2 (Web2C 7.5.4): "Image scaling is presently implemented only in TeX, not in HTML or any other sort of output." -- Gary Cramblitt (aka PhantomsDad) KDE Text-to-Speech Maintainer http://accessibility.kde.org/developer/kttsd/index.php _______________________________________________ Speechd mailing list Speechd@lists.freebsoft.org http://lists.freebsoft.org/mailman/listinfo/speechd From mooncar@libero.it Wed Mar 15 11:23:57 2006 From: mooncar@libero.it To: speechd-discuss@nongnu.org Date: Sun Aug 13 12:52:51 2006 Subject: shell under emacs Message-ID: Thank you to all for your help! About using shell, I know there is term too, but it's not a problem, this person is using emacs at university and there are many professors who are helping him and know what is useful and what is not, I don't know what he must do-it study phisics- thanks you again! vero > > You can get the previous command history by using C-p and the next one > by using C-n when running M-x shell. However, note that you probably > better off using M-x term rather than M-x shell. The 'term' mode is a > lot more sophisticated and robust for shell style interaction. the > 'shell' command is very basic and really only useful for very simple > 'quick and dirty' commands rather than for more extended shell > interaction. Another very useful shell is M-x eshell, which is a shell > written in emacs lisp. The nice aspect of this shell is that you have > full intergration with emacs and can use commands like C-x C-f to find > a file without having to use terminal escape characters or different > key bindings that don't interfere with the shell etc. > > I find it rare that you actually need to use the shell. This is a > common misconception with new users. I would strongly recommend > posting some description of what this user is using the shell for as > it is very likely there are alternative solutions which work even > better with emacspeak and provide superior speech feedback. For > example, using dired rather than ls to get a listing of what is in a > directory and using tramp to access files on remote systems etc. > > HTH > > Tim > mooncar@libero.it writes: > > > > Hi! > > I have installed emacs with speech support,now it works well,but the boy > > who is using that tells me he needs to use the command M-x shell, in emacs. > > Is it possible to recall the commands that he have just issued, > > like in the linux shell out of emacs?now he must re-type the commands... > > I have tryied to find a solution but I don't use that program-I only > > installed and tryied if it's working- so if someone can help... > > thanks in advance, > > veronica > > > > _______________________________________________ > > Speechd mailing list > > Speechd@lists.freebsoft.org > > http://lists.freebsoft.org/mailman/listinfo/speechd > > ----------------------------------------------------------------------------- > To unsubscribe from the emacspeak list or change your address on the > emacspeak list send mail to "emacspeak-request@cs.vassar.edu" with a > subject of "unsubscribe" or "help" > > _______________________________________________ Speechd mailing list Speechd@lists.freebsoft.org http://lists.freebsoft.org/mailman/listinfo/speechd From mooncar@libero.it Tue Mar 14 15:09:27 2006 From: mooncar@libero.it To: speechd-discuss@nongnu.org Date: Sun Aug 13 12:52:51 2006 Subject: shell under emacs Message-ID: Hi! I have installed emacs with speech support,now it works well,but the boy who is using that tells me he needs to use the command M-x shell, in emacs. Is it possible to recall the commands that he have just issued, like in the linux shell out of emacs?now he must re-type the commands... I have tryied to find a solution but I don't use that program-I only installed and tryied if it's working- so if someone can help... thanks in advance, veronica _______________________________________________ Speechd mailing list Speechd@lists.freebsoft.org http://lists.freebsoft.org/mailman/listinfo/speechd From mooncar@libero.it Thu Feb 16 18:15:41 2006 From: mooncar@libero.it To: speechd-discuss@nongnu.org Date: Sun Aug 13 12:52:51 2006 Subject: error while loading shared libraries: libdotconf Message-ID: I have resolved by adding /usr/local/lib in the conf file for ldconfig, and re-running ldconfig. Now i haven't errors when i start speechdispatcher, but when I type spd-say "hello" the connection is closed.. the festival.log file is Thu Feb 16 17:44:39 2006 [509392] ALSA: Opening ALSA sound output Thu Feb 16 17:44:39 2006 [509466] ALSA: Opening ALSA device Thu Feb 16 17:44:39 2006 [511380] ALSA: Allocating new sw_params structure Thu Feb 16 17:44:39 2006 [511422] ALSA: Opening ALSA device ... success Thu Feb 16 17:44:39 2006 [511435] ALSA: Device 'default' initialized succesfully. Thu Feb 16 17:45:02 2006 [784243] ALSA: Start of playback on ALSA Thu Feb 16 17:45:02 2006 [784298] ALSA: Allocating new hw_params structure Thu Feb 16 17:45:02 2006 [784894] ALSA: PCM state before setting audio parameters: OPEN Thu Feb 16 17:45:02 2006 [784921] ALSA: Setting access type to INTERLEAVED Thu Feb 16 17:45:02 2006 [785054] ALSA: Setting sample format to S16_LE Thu Feb 16 17:45:02 2006 [785135] ALSA: Setting sample rate to 16000 Thu Feb 16 17:45:02 2006 [785272] ALSA: Setting channel count to 1 Thu Feb 16 17:45:02 2006 [785330] ALSA: Setting hardware parameters on the ALSA device Thu Feb 16 17:45:02 2006 [786020] ALSA: Buffer size on ALSA device is 677 bytes Thu Feb 16 17:45:02 2006 [786044] ALSA: Preparing device for playback Thu Feb 16 17:45:02 2006 [786157] ALSA: volume size = 25800 Thu Feb 16 17:45:02 2006 [786169] ALSA: Making copy of track and adjusting volume Thu Feb 16 17:45:02 2006 [786326] ALSA: Filling with silence up to the period size, silent_samples=1 pcm.c:5958: snd_pcm_mmap_commit: Assertion `frames <= snd_pcm_mmap_avail(pcm)' failed. My festival is ok, it works... it's festival 1.4.2. > > mi> I have just installed speech dispatcher 0.6 from tarball, the > > mi> installation was successfully but when I run speech-dispatcher I > > mi> have this error > > mi> speech-dispatcher: error while loading shared libraries: > > mi> libdotconf-1.0.so.0: cannot open shared object file: No such > > mi> file or directory > > Try install speechd again but before remove all libspeechd. I'm not sure > > if it help.... > > No, I don't think this will help. It has nothing to do with libspeechd > and removing all libspeechd is not a good idea because some applications > on the system might be linked against older versions of libspeechd > and this would break them. > > mooncar, I think there is some mess with the instalation of libdotconf. > The configure script for Dispatcher performs a check if the library > is available and from your writing it seems it passed the check. > You also say you have libdotconf installed. Maybe you did configure > as another user as the one who is trying to do make or something > like that? > > I'd guess that gcc is not looking into /usr/local/lib for the library > and only looks into /usr/lib. Try compiling again, starting with > make clean > LDFLAGS="-L/usr/local/lib" ./configure > > If it doesn't help, do > > make clean > ./configure 2&> configure.log > make 2&> make.log > > and send me the two .log files so that I can see what is wrong. > > Have a nice day, > Hynek > > _______________________________________________ Speechd mailing list Speechd@lists.freebsoft.org http://lists.freebsoft.org/mailman/listinfo/speechd From tcross@rapttech.com.au Wed Mar 15 09:01:18 2006 From: Tim Cross To: speechd-discuss@nongnu.org Date: Sun Aug 13 12:52:51 2006 Subject: shell under emacs In-Reply-To: References: Message-ID: <17431.51662.617759.229441@tiger.rapttech.com.au> You can get the previous command history by using C-p and the next one by using C-n when running M-x shell. However, note that you probably better off using M-x term rather than M-x shell. The 'term' mode is a lot more sophisticated and robust for shell style interaction. the 'shell' command is very basic and really only useful for very simple 'quick and dirty' commands rather than for more extended shell interaction. Another very useful shell is M-x eshell, which is a shell written in emacs lisp. The nice aspect of this shell is that you have full intergration with emacs and can use commands like C-x C-f to find a file without having to use terminal escape characters or different key bindings that don't interfere with the shell etc. I find it rare that you actually need to use the shell. This is a common misconception with new users. I would strongly recommend posting some description of what this user is using the shell for as it is very likely there are alternative solutions which work even better with emacspeak and provide superior speech feedback. For example, using dired rather than ls to get a listing of what is in a directory and using tramp to access files on remote systems etc. HTH Tim mooncar@libero.it writes: > > Hi! > I have installed emacs with speech support,now it works well,but the boy > who is using that tells me he needs to use the command M-x shell, in emacs. > Is it possible to recall the commands that he have just issued, > like in the linux shell out of emacs?now he must re-type the commands... > I have tryied to find a solution but I don't use that program-I only > installed and tryied if it's working- so if someone can help... > thanks in advance, > veronica > > _______________________________________________ > Speechd mailing list > Speechd@lists.freebsoft.org > http://lists.freebsoft.org/mailman/listinfo/speechd _______________________________________________ Speechd mailing list Speechd@lists.freebsoft.org http://lists.freebsoft.org/mailman/listinfo/speechd From mooncar@libero.it Fri Jan 20 17:27:13 2006 From: mooncar@libero.it To: speechd-discuss@nongnu.org Date: Sun Aug 13 12:52:51 2006 Subject: Speechd-el in italian Message-ID: Yes now, with recode utility, emacs speaks in Italian! THANK YOU !!! Best wishes, veronica _______________________________________________ Speechd mailing list Speechd@lists.freebsoft.org http://lists.freebsoft.org/mailman/listinfo/speechd From garycramblitt@comcast.net Thu Mar 30 04:27:03 2006 From: Gary Cramblitt To: speechd-discuss@nongnu.org Date: Sun Aug 13 12:52:51 2006 Subject: Documentation Message-ID: <200603292127.03239.garycramblitt@comcast.net> I have committed spelling and grammar corrections for speech-dispatcher.texi. I encountered a couple of issues. 1. When I do "make html", I get this: cd .. && /bin/sh /home/share/src/speechd/missing --run automake-1.9 --gnu doc/Makefile cd .. && /bin/sh ./config.status doc/Makefile config.status: creating doc/Makefile make: *** No rule to make target `speechd.html', needed by `html'. Stop. When I modify the html_docs line in doc/Makefile.am from html_docs = speechd.html ssip.html spd-say.texi to html_docs = speech-dispatcher.html ssip.html spd-say.texi then I do not get the error. 2. When generating a pdf of the manual, the figures/architecture.pdf file is too wide for the page. This causes a blank area at the point where the file is referenced and part of the image appears a page or two later. I committed a smaller version of architecture.pdf to fix this problem, but it is quite fuzzy. I'd be grateful if someone could figure out how to scale the image down but retain sharpness. -- Gary Cramblitt (aka PhantomsDad) _______________________________________________ Speechd mailing list Speechd@lists.freebsoft.org http://lists.freebsoft.org/mailman/listinfo/speechd From mooncar@libero.it Mon Jan 23 21:25:03 2006 From: mooncar@libero.it To: speechd-discuss@nongnu.org Date: Sun Aug 13 12:52:51 2006 Subject: Need help with festival in red hat Message-ID: Hello! I have to install speechd with festival on a computer with Linux Red Hat 8. I have already installed FESTIVAL by tarball, but it speak too fast. The exact procedure that I have followed is tar xvzf speech_tools-1.2.95.tar.gz ./configure, make, make install tar xvzf festival-1.95-beta.tar.gz tar xvzf festlex_CMU.tar.gz tar xvzf festlex_OALD.tar.gz tar xvzf festlex_POSLEX.tar.gz tar xvzf festvox_kallpc16k.tar.gz ./configure, make, make install It seems that all goes well, but when I run Festival it speak too fast, and if I use the command (Parameter.set 'Duration_Stretch 2) the result is that pronounciation in stretched, but still deformed.. Why? Has someone any idea ? thank you mooncar _______________________________________________ Speechd mailing list Speechd@lists.freebsoft.org http://lists.freebsoft.org/mailman/listinfo/speechd From nath.ml@free.fr Mon Feb 27 20:00:55 2006 From: Nath To: speechd-discuss@nongnu.org Date: Sun Aug 13 12:52:51 2006 Subject: problem with w3 and w3m In-Reply-To: (mooncar@libero.it's message of "Mon, 27 Feb 2006 17:02:23 +0100") References: Message-ID: <87fym4pqxk.fsf@free.fr> "mooncar\@libero\.it" writes: > Done, but during the ./configure script I have this > > checking GC library exists... yes > checking gc.h usability... no > checking gc.h presence... no > checking for gc.h... no > checking GC header location... /usr /usr/local /root > checking /usr/include... checking gc.h usability... no > [...] > checking for gc.h... no > configure: error: gc.h not found > > Maybe I've missing something? > > Hello, Which distribution are you using ? If debian try to install libgc-dev Perhaps it will help ? -- Nath _______________________________________________ Speechd mailing list Speechd@lists.freebsoft.org http://lists.freebsoft.org/mailman/listinfo/speechd From garycramblitt@comcast.net Thu Feb 16 23:04:52 2006 From: Gary Cramblitt To: speechd-discuss@nongnu.org Date: Sun Aug 13 12:52:51 2006 Subject: error while loading shared libraries: libdotconf In-Reply-To: References: Message-ID: <200602161704.52248.garycramblitt@comcast.net> On Thursday 16 February 2006 12:15, mooncar@libero.it wrote: > Buffer size on ALSA device is 677 bytes > Thu Feb 16 17:45:02 2006 [786044] ALSA: Preparing device for playback > Thu Feb 16 17:45:02 2006 [786157] ALSA: volume size = 25800 These numbers look a bit suspicious to me. Try using another ALSA pcm device. In the festival.conf file change the line to FestivalALSADevice "hw:0,0" or FestivalALSADevice "plughw:0,0" Try playing a wav file using aplay aplay somefile.wav If this doesn't help we may have to uncomment some of the debug statements in the alsa.c code to see what is going on. -- Gary Cramblitt (aka PhantomsDad) _______________________________________________ Speechd mailing list Speechd@lists.freebsoft.org http://lists.freebsoft.org/mailman/listinfo/speechd From cerha@brailcom.org Fri Mar 31 07:37:19 2006 From: Tomas Cerha To: speechd-discuss@nongnu.org Date: Sun Aug 13 12:52:51 2006 Subject: Documentation In-Reply-To: <200603301845.20522.garycramblitt@comcast.net> References: <200603292127.03239.garycramblitt@comcast.net> <1143753979.4238.38.camel@chopin> <200603301845.20522.garycramblitt@comcast.net> Message-ID: <442CC00F.6030207@brailcom.org> Gary Cramblitt napsal(a): > Afterall dia is a vector based image, so it should be possible to get a > crystal-clear image. Needs to be pdf of a image about 450x155 pixels if > anyone wants to give it a go. Hi, I produced all the previous formats from the original dia drawing, so I believe I'll be able to do that again. I don't remember how I did it, but I'll give it a try... Best regards Tomas Cerha -- Brailcom, o.p.s. http://www.brailcom.org Free(b)soft project http://www.freebsoft.org Eurochance project http://eurochance.brailcom.org _______________________________________________ Speechd mailing list Speechd@lists.freebsoft.org http://lists.freebsoft.org/mailman/listinfo/speechd From buchal@brailcom.org Tue Jan 24 12:36:14 2006 From: Jan Buchal To: speechd-discuss@nongnu.org Date: Sun Aug 13 12:52:51 2006 Subject: Need help with festival in red hat In-Reply-To: <864q3tdfva.fsf@brailcom.org> (buchal@brailcom.org's message of "Tue, 24 Jan 2006 12:28:41 +0100") References: <864q3tdfva.fsf@brailcom.org> Message-ID: <86acdlc0y9.fsf@brailcom.org> >>>>> "b" == buchal writes: >>>>> "mi" == mooncar\@libero\ it writes: b> p.s. This email come on my personal email only. I'm not sure if b> you wanted it. Sorry, I'm wrong. the email come in mailing list too. -- Jan Buchal Tel: (00420) 24 24 86 008 Mob: (00420) 608023021 _______________________________________________ Speechd mailing list Speechd@lists.freebsoft.org http://lists.freebsoft.org/mailman/listinfo/speechd From cerha@brailcom.org Fri Mar 31 10:19:30 2006 From: Tomas Cerha To: speechd-discuss@nongnu.org Date: Sun Aug 13 12:52:51 2006 Subject: Documentation In-Reply-To: <442CC00F.6030207@brailcom.org> References: <200603292127.03239.garycramblitt@comcast.net> <1143753979.4238.38.camel@chopin> <200603301845.20522.garycramblitt@comcast.net> <442CC00F.6030207@brailcom.org> Message-ID: <442CE612.8020503@brailcom.org> Tomas Cerha wrote: > Hi, I produced all the previous formats from the original dia drawing, > so I believe I'll be able to do that again. I don't remember how I did > it, but I'll give it a try... Hello, the solution was quite simple at the end. I simply converted the EPS to a vector graphics PDF using epstopdf and added an explicit image width 155mm to the texi source file. The image is automatically scaled by TeX to fit the given width and the result remains a vector graphics, so it will retain the full resolution when printed. Best regards Tomas -- Brailcom, o.p.s. http://www.brailcom.org Free(b)soft project http://www.freebsoft.org Eurochance project http://eurochance.brailcom.org _______________________________________________ Speechd mailing list Speechd@lists.freebsoft.org http://lists.freebsoft.org/mailman/listinfo/speechd From garycramblitt@comcast.net Tue Mar 28 04:11:22 2006 From: Gary Cramblitt To: speechd-discuss@nongnu.org Date: Sun Aug 13 12:52:51 2006 Subject: [Documentation] ModuleDelimiters Message-ID: <200603272111.22845.garycramblitt@comcast.net> I'm working on the Speech Dispatcher documentation. The section on module configuration claims that the Festival module breaks text up into sentences and parts of sentences according to ModuleDelimiters and ModuleMaxChunkLengh parameters. Is this true anymore for the Festival module? I looks like it is true for Flite module. Thanks -- Gary Cramblitt (aka PhantomsDad) _______________________________________________ Speechd mailing list Speechd@lists.freebsoft.org http://lists.freebsoft.org/mailman/listinfo/speechd From mooncar@libero.it Fri Jan 20 11:54:11 2006 From: mooncar@libero.it To: speechd-discuss@nongnu.org Date: Sun Aug 13 12:52:51 2006 Subject: Speechd in Italian Message-ID: Hi to all! I am new with speech dispatcher, I have just installed it because I need emacs with Italian speech. But I have some problems... I have Debian kanotix and emacs 21 (and emacs21-el), and I have installed festival 1.4.3 mbrola with italian voices italian festival by CNR ISTC by Piero Cosi libdotconf festival-freebsoft-utils 0.4-1 speech-dispatcher 0.5-4 speech-dispatcher-festival 0.5-4 libspeechd 0.5-4 speechd-el 1.0-3 I have edited the file /etc/speech-dispatcher/speechd.conf writing AddModule "festival" "sd_festival" "festival.conf" DefaultModule festival LanguageDefaultModule "en" "festival" And I have added (it italian) to the variable language-codes at the beginning of the freebsoft utils file voice-select.scm. I start festival --server in a shell, and in another shell I become root and start speech-dispatcher. The connection is ok: server Thu Jan 19 17:55:27 2006 : Festival server started on port 1314 client(1) Thu Jan 19 17:55:32 2006 : accepted from curaro If I start emacs and use the command M-x speechd-speak, I have in the festival shell: SIOD ERROR: could not open file /usr/lib/festival/synthesis.scm SIOD: unknown voice kal_diphone SIOD ERROR: could not open file /usr/lib/festival/synthesis.scm SIOD: unknown voice lp_diphone SIOD ERROR: could not open file /usr/lib/festival/synthesis.scm SIOD: unknown voice pc_diphone SIOD ERROR: could not open file /usr/lib/festival/synthesis.scm SIOD: unknown voice pc_mbrola SIOD ERROR: could not open file /usr/lib/festival/synthesis.scm SIOD ERROR: could not open file /usr/lib/festival/synthesis.scm SIOD ERROR: wrong type of argument to times SIOD ERROR: could not open file /usr/lib/festival/synthesis.scm sh: recode: command not found SIOD ERROR: could not open file /usr/lib/festival/synthesis.scm sh: recode: command not found SIOD ERROR: wrong type of argument to get_c_val sh: recode: command not found SIOD ERROR: wrong type of argument to get_c_val SIOD ERROR: wrong type of argument to get_c_utt sh: recode: command not found Phoneme: pau has no duration info SIOD ERROR: wrong type of argument to get_c_utt sh: recode: command not found Phoneme: pau has no duration info SIOD ERROR: wrong type of argument to get_c_utt SIOD ERROR: could not open file /usr/lib/festival/synthesis.scm sh: recode: command not found SIOD ERROR: could not open file /usr/lib/festival/synthesis.scm sh: recode: command not found SIOD ERROR: wrong type of argument to get_c_val Because in my system all festival files are in the folder named /usr/share/festival, I copied the file 'synthesis.scm' from /usr/share/festival to /usr/lib/festival. Now, I have emacs speaking in English. I use the command M-x speechd-set-language,I answer it, and emacs become silent, it no longer speak. the festival shell say: sh: recode: command not found sh: recode: command not found sh: recode: command not found sh: recode: command not found sh: recode: command not found SIOD ERROR: wrong type of argument to times sh: recode: command not found sh: recode: command not found sh: recode: command not found sh: recode: command not found And the log file: [Thu Jan 19 18:18:08 2006 : 841130] speechd: Speech Dispatcher Lo gging to file /var/log/speech-dispatcher/speech-dispatcher.log [Thu Jan 19 18:18:08 2006 : 909868] speechd: Module festival loaded. [Thu Jan 19 18:18:08 2006 : 989181] speechd: Module festival started sucessfully with message: --------------- Communicating with Festival through a socket. --------------- [Thu Jan 19 18:18:08 2006 : 989375] speechd: Configuration has been read from "/etc/speech-dispatcher //speechd.conf" [Thu Jan 19 18:18:08 2006 : 989408] speechd: Speech Dispatcher started with 1 output module [Thu Jan 19 18:18:08 2006 : 989516] speechd: Openning socket connection [Thu Jan 19 18:18:08 2006 : 989584] speechd: Speech Dispatcher waiting for clients ... [Thu Jan 19 18:18:08 2006 : 989584] speechd: Speech Dispatcher waiting for clients ... [Thu Jan 19 18:18:37 2006 : 966861] speechd: Error: Failed to serve client on fd 7! [Thu Jan 19 18:18:37 2006 : 967606] speechd: Error: Failed to serve client on fd 7! [Thu Jan 19 18:18:37 2006 : 967868] speechd: Error: Failed to serve client on fd 7! [Thu Jan 19 18:18:37 2006 : 968012] speechd: Error: Failed to serve client on fd 7! [Thu Jan 19 18:18:37 2006 : 968129] speechd: Connection closed [Thu Jan 19 18:19:07 2006 : 381209] speechd: Error: Failed to serve client on fd 10! [Thu Jan 19 18:19:07 2006 : 387093] speechd: Error: Failed to serve client on fd 10! [Thu Jan 19 18:19:07 2006 : 387481] speechd: Error: Failed to serve client on fd 10! [Thu Jan 19 18:19:07 2006 : 387538] speechd: Error: Failed to serve client on fd 10! [Thu Jan 19 18:19:07 2006 : 387591] speechd: Connection closed [Thu Jan 19 18:19:10 2006 : 532412] speechd: Error: Failed to serve client on fd 7! [Thu Jan 19 18:19:10 2006 : 533002] speechd: Error: Failed to serve client on fd 7! [Thu Jan 19 18:19:10 2006 : 533059] speechd: Error: Failed to serve client on fd 7! [Thu Jan 19 18:19:10 2006 : 533106] speechd: Connection closed someone can help me to understand what it is happening? Thank you in advance veronica _______________________________________________ Speechd mailing list Speechd@lists.freebsoft.org http://lists.freebsoft.org/mailman/listinfo/speechd From garycramblitt@comcast.net Sun May 7 23:02:53 2006 From: Gary Cramblitt To: speechd-discuss@nongnu.org Date: Sun Aug 13 12:52:49 2006 Subject: Audio logging Message-ID: <200605071702.53145.garycramblitt@comcast.net> I noticed that the audio output routines, alsa, oss, and nas all output to the log files even if Debug is 0 in the output module conf file. Anyone object if I modify to only log actual errors if Debug = 0? -- Gary Cramblitt (aka PhantomsDad) _______________________________________________ Speechd mailing list Speechd@lists.freebsoft.org http://lists.freebsoft.org/mailman/listinfo/speechd From garycramblitt@comcast.net Fri May 12 01:25:18 2006 From: Gary Cramblitt To: speechd-discuss@nongnu.org Date: Sun Aug 13 12:52:49 2006 Subject: [Announce] Speech Dispatcher IBM TTS Output Module ready for testing Message-ID: <200605111925.18959.garycramblitt@comcast.net> A new output module in Speech Dispatcher is ready for user testing. The ibmtts output module uses the commercial IBM TTS synthesizer for Linux and offers the following capabilities: 1. Support for all the documented dialects and voices of IBM TTS, plus custom voices via the ibmtts.conf file. 2. Support for spelling mode. 3. Support for changing volume, pitch, rate, etc. 4. Support for Speech Synthesis Markup Language (SSML) (as far as IBM TTS supports it). 5. Support for index marking, including SSML tags. 6. Support for pronouncing all the Speech Dispatcher KEY commands (English, but can be customized via the ibmtts.conf file). 7. Support for all the characters in the Latin-1 charset (English, but can be customized via the ibmtts.conf file). 8. Support for sound icons via separate .wav files. In order to test the new ibmtts output module, you will need: 1. A copy of the commercial IBM TTS synthesizer for Linux. More information available at http://ibmtts-sdk.sourceforge.net/ 2. Ability to download and build Speech Dispatcher from CVS. More info at http://www.freebsoft.org/speechd I'm interested in feedback on any compile/build problems that users might have as well as usability. If anyone develops enhancements for the ibmtts.conf file for other languages, please send them to me. If you do not have a copy of the IBM TTS synthesizer itself, you can still test building of the ibmtts output module by downloading the IBM TTS Software Development Kit (SDK), which is available at the SourceForge link given above. You must retrieve the SDK using SourceForge CVS. See README.packagers file in the Speech Dispatcher source. Without the actual synthesizer, you won't hear anything, of course. The SDK (not the synthesizer) is licensed under a BSD-style license. In addition, the latest CVS version of Speech Dispatcher includes a few enhancements, including 1. Updated documentation. 2. An --ssml option for spd-say for speaking SSML. 3. A --pipe-mode option for spd-say that permits inserting spd-say into a Unix stdin/stdout pipe. See the spd-say manual for more information. 4. Additional test scripts in the src/tests directory. 5. Enhanced espeak-generic.conf file for use with ESpeak. The Speech Dispatcher project provides a device independent layer for speech synthesis through a simple, stable and well documented interface. It takes care of most of the tasks necessary to solve in speech enabled applications. What is a very high level GUI library to graphics, Speech Dispatcher is to speech synthesis. Speech Dispatcher provides excellent support for the Free Festival speech synthesizer, as well as Festival Lite (flite), ESpeak (via generic module), Epos, DecTalk, Apollo, llia, and (now) IBM TTS. -- Gary Cramblitt (aka PhantomsDad) _______________________________________________ Speechd mailing list Speechd@lists.freebsoft.org http://lists.freebsoft.org/mailman/listinfo/speechd From garycramblitt@comcast.net Sun May 7 23:37:25 2006 From: Gary Cramblitt To: speechd-discuss@nongnu.org Date: Sun Aug 13 12:52:49 2006 Subject: Default Volume In-Reply-To: <200605071638.34980.garycramblitt@comcast.net> References: <200605071638.34980.garycramblitt@comcast.net> Message-ID: <200605071737.25819.garycramblitt@comcast.net> On Sunday 07 May 2006 16:38, Gary Cramblitt wrote: > According to the SSIP manual, the default volume is 0 (range -100 to 100). > I don't think that is correct. When I run spd-say, the default volume is > 100. Nevermind, just saw this in speech.conf file: # DefaultVolume constrols the default volume of the voice. # It is a value between -100 (softly) and +100 (loudly). # Currently, +100 maps to the default volume of the synthesizer. DefaultVolume 100 -- Gary Cramblitt (aka PhantomsDad) _______________________________________________ Speechd mailing list Speechd@lists.freebsoft.org http://lists.freebsoft.org/mailman/listinfo/speechd From garycramblitt@comcast.net Sun May 7 23:02:53 2006 From: Gary Cramblitt To: speechd-discuss@nongnu.org Date: Sun Aug 13 12:52:49 2006 Subject: Audio logging Message-ID: <200605071702.53145.garycramblitt@comcast.net> I noticed that the audio output routines, alsa, oss, and nas all output to the log files even if Debug is 0 in the output module conf file. Anyone object if I modify to only log actual errors if Debug = 0? -- Gary Cramblitt (aka PhantomsDad) _______________________________________________ Speechd mailing list Speechd@lists.freebsoft.org http://lists.freebsoft.org/mailman/listinfo/speechd From jsd@clara.co.uk Sun Apr 30 14:24:58 2006 From: Jonathan Duddington To: speechd-discuss@nongnu.org Date: Sun Aug 13 12:52:49 2006 Subject: eSpeak - added punctuation and capitals indications In-Reply-To: <200604300724.23083.garycramblitt@comcast.net> References: <4e1b823a18jsd@clara.co.uk> <200604300724.23083.garycramblitt@comcast.net> Message-ID: <4e1facd663jsd@clara.co.uk> In article <200604300724.23083.garycramblitt@comcast.net>, Gary Cramblitt wrote: > The main problem I noticed is that you do not provide a library or > api for interfacing directly with espeak. Everything is done through > command line and manipulation of configuration files.... > You are aware, I believe, that we are currently discussing a new TTS > Engine API on accessibility@freedesktop.org. Hynek and I are hoping > that this new API will be approved soon, whereupon we will begin a > major refactoring of Speech Dispatcher to use this api. Therefore, > it would be best if you could design an api for espeak that would be > aligned with that specification. Yes. I intend to produce a version of eSpeak as a library, with direct function calls from a Speech Dispatcher driver. It would not include the portaudio interface, but would return synthesized speech in memory buffers, through a function callback. I'm hoping to produce a proposed interface specification for the eSpeak Library within a few days, which will take into account the latest (28.4.06) Common TTS API draft. > Some of the major changes I would like to see in espeak are: > 1. Support for SSML. I noticed that you now support embedded > commands for controlling rate, pitch, volume, etc. In order to use > these, SD would have to parse the SSML itself and translate embedded > SSML into your command syntax. That would be inefficient and > probably imperfect. It would be better if espeak directly supported > SSML. The embedded commands (and the UTF-8 capability) are a preparation for this. The SSML tags need to be broken down into the lower level commands to change pitch, rate, etc. I had thought that Speech Dispatcher needs to do this for synthesizers which don't process SSML themselves, and that I might also be able to take advantage of S.D. ability to this :-) But if you think it's better for eSpeak to process SSML directly, then that's OK, subject to a couple of problems which I'll address with the interface proposal. > 2. Espeak needs to return audio directly to SD. Writing a .wav file > to disk is inefficient. A more direct method, such as callbacks or > socket io would be better. Yes, see above. > 3. Support for index marking. Ideally, espeak should provide > callbacks and/or index mark information for the following: > a. Begin/end of entire message. > b. Begin/end of sentence. > c. Begin/end of word. > d. Custom index marks as tags in SSML. Yes, eSpeak's callbacks would return both synthesized speech data and index data to S.D. The RISC OS version of eSpeak does currently produces word index information which an application uses to move a caret through the text and scroll it as it's spoken. I hope to produce a library version of eSpeak soon, with basic features (i.e. those equivalent to the current command-line version), so it would be better to wait for that rather than doing any work on a S.D. driver for the command-line version. -- _______________________________________________ Speechd mailing list Speechd@lists.freebsoft.org http://lists.freebsoft.org/mailman/listinfo/speechd From cerha@brailcom.org Wed Jun 28 18:06:16 2006 From: Tomas Cerha To: speechd-discuss@nongnu.org Date: Sun Aug 13 12:52:49 2006 Subject: a new module in speech-dispatcher In-Reply-To: <20060622220521.GA27784@sunset.terramitica.net> References: <20060622220521.GA27784@sunset.terramitica.net> Message-ID: <44A2A8F8.9010507@brailcom.org> Olivier BERT wrote: > I added a new output module to speech-dispatcher. > It handles the cicero TTS software. Hi Olivier, thank you for your effort. What is the current status of the module? Is it stable enough, so that we can consider releasing a new Speech Dispatcher version? Best regards, Tomas. -- Brailcom, o.p.s. http://www.brailcom.org Free(b)soft project http://www.freebsoft.org Eurochance project http://eurochance.brailcom.org _______________________________________________ Speechd mailing list Speechd@lists.freebsoft.org http://lists.freebsoft.org/mailman/listinfo/speechd From cerha@brailcom.org Sat Apr 1 09:13:03 2006 From: Tomas Cerha To: speechd-discuss@nongnu.org Date: Sun Aug 13 12:52:49 2006 Subject: Documentation In-Reply-To: <200603311644.58103.garycramblitt@comcast.net> References: <200603292127.03239.garycramblitt@comcast.net> <442CC00F.6030207@brailcom.org> <442CE612.8020503@brailcom.org> <200603311644.58103.garycramblitt@comcast.net> Message-ID: <442E27FF.5050704@brailcom.org> Gary Cramblitt wrote: > I looked for an eps2pdf program but could only find them for Windows and I'm > running Debian Linux. Could you convert the file for me and send it? It > might work better for me. Hi Gary, I committed the pdf file into the CVS. The program epstopdf is a part of the tetex-bin Debian package. > I can use ImageMagick to convert an eps file to pdf, but when I add 155mm to > the texi source as you suggest, it is ignored. The result is only part of > the image fits on the page and texinfo mangles the document pretty badly. > According to my texinfo manual, Version 3.141592-1.21a-2.2 (Web2C 7.5.4): > > "Image scaling is presently implemented only in TeX, not in HTML or any other > sort of output." This should be ok, I guess, since the PDF manual is created using TeX. The HTML output uses the PNG image, which is scaled. The info output uses the ascii version. I'm using the Debian texinfo package version 4.8-4. I'd try running texinfo with the PDF figure from CVS. If that doesn't help, the problem might be in your version of texinfo. Hope that helps. Best regards Tomas -- Brailcom, o.p.s. http://www.brailcom.org Free(b)soft project http://www.freebsoft.org Eurochance project http://eurochance.brailcom.org _______________________________________________ Speechd mailing list Speechd@lists.freebsoft.org http://lists.freebsoft.org/mailman/listinfo/speechd From hanke@brailcom.org Thu May 11 21:24:11 2006 From: Hynek Hanke To: speechd-discuss@nongnu.org Date: Sun Aug 13 12:52:49 2006 Subject: Audio logging In-Reply-To: <200605071702.53145.garycramblitt@comcast.net> References: <200605071702.53145.garycramblitt@comcast.net> Message-ID: <1147375451.4386.2.camel@chopin> Gary Cramblitt p??e v Ne 07. 05. 2006 v 17:02 -0400: > I noticed that the audio output routines, alsa, oss, and nas all output to the > log files even if Debug is 0 in the output module conf file. Anyone object > if I modify to only log actual errors if Debug = 0? No. With regards, Hynek Hanke _______________________________________________ Speechd mailing list Speechd@lists.freebsoft.org http://lists.freebsoft.org/mailman/listinfo/speechd From garycramblitt@comcast.net Sat May 13 15:57:45 2006 From: Gary Cramblitt To: speechd-discuss@nongnu.org Date: Sun Aug 13 12:52:49 2006 Subject: [Announce] Speech Dispatcher IBM TTS Output Module ready for testing In-Reply-To: <200605111925.18959.garycramblitt@comcast.net> References: <200605111925.18959.garycramblitt@comcast.net> Message-ID: <200605130957.45803.garycramblitt@comcast.net> On Thursday 11 May 2006 19:25, Gary Cramblitt wrote: > A new output module in Speech Dispatcher is ready for user testing. The > ibmtts output module uses the commercial IBM TTS synthesizer for Linux and > offers the following capabilities: Just a quick note that SourceForge just changed the procedures for using their CVS. The commands to get the IBM TTS SDK would be: cvs -d:pserver:anonymous@ibmtts-sdk.cvs.sourceforge.net:/cvsroot/ibmtts-sdk login cvs -d:pserver:anonymous@ibmtts-sdk.cvs.sourceforge.net:/cvsroot/ibmtts-sdk co sdk -- Gary Cramblitt (aka PhantomsDad) _______________________________________________ Speakup mailing list Speakup@braille.uwo.ca http://speech.braille.uwo.ca/mailman/listinfo/speakup From hanke@brailcom.org Thu May 11 21:48:18 2006 From: Hynek Hanke To: speechd-discuss@nongnu.org Date: Sun Aug 13 12:52:49 2006 Subject: Default Volume In-Reply-To: <200605071744.01001.garycramblitt@comcast.net> References: <200605071638.34980.garycramblitt@comcast.net> <200605071737.25819.garycramblitt@comcast.net> <200605071744.01001.garycramblitt@comcast.net> Message-ID: <1147376898.4386.19.camel@chopin> Hi Gary, DefaultModule being set to 100 in the configuration is a hack to deal with the current limitation of our output modules. The idea in SSIP to have 0 for default, with the possibility to make the sound louder through positive values, is correct, I'd say. However, the current sound output modules only achieve volume modification through multiplying the sample data with a constant factor. Of course, this factor needs to be less than one, otherwise the samples can overflow. I do not currently know how to make samples comming from a synthesizer louder with the simple audio output we use. Maybe there is a way through ALSA? Until this gets fixed, SSIP value 0 for volume will have the effect of 50% sound volume level comming from the synthesizer and thus we need to have 100 as default in speechd.conf. I however think we should not spoil clients, their documentation, SSIP etc. with this implementation detail of Dispatcher. spd-say is doing the correct thing, unless the SSIP documentation is changed. Of course I'm open to suggestions. With regards, Hynek _______________________________________________ Speechd mailing list Speechd@lists.freebsoft.org http://lists.freebsoft.org/mailman/listinfo/speechd From jsd@clara.co.uk Sat Apr 22 12:14:44 2006 From: Jonathan Duddington To: speechd-discuss@nongnu.org Date: Sun Aug 13 12:52:49 2006 Subject: eSpeak - added punctuation and capitals indications Message-ID: <4e1b823a18jsd@clara.co.uk> eSpeak is a compact open-source text-to-speech engine which I hope to adapt to work well with Speech Dispatcher. I've added the ability to indicate punctuation and capital letters in the text by either speaking their name, or by playing a sound. I'm not sure whether what I've done is exactly what's wanted, so if anyone wants to try it and let me know of suggestions for improvements, please do so. eSpeak text-to-speech is at: http://espeak.sourceforge.net download the file: test-1.09d-linux.zip The ReadMe file inside gives the details. It can speak either all punctuation or just a specified set of punctuation characters. It can speak their names, or you can set up sound files (sound icons) to be played instead. Capital letters can be indicated by a sound, or by the word "capital", or by raising the pitch of the capitalized word. Also the feature of embedding commands within the text has been updated. Questions which I'm unsure about: 1. Should end-of-line be indicated? 2. What about apostrophes within words. Currently these are not indicated when speaking text since that would disrupt the pronunciation of the word. 3. The punctuation name is spoken in a slightly different tone from the main text, to differentiate it. Is that OK? 4. The actual names for punctuation characters are defined in the data/english_list file, so these can be changed if needed (then do speak --compile). 5. If the text is spoken at a fast rate, should the sound icons also be shortened in duration? 6. What is the best value for the pitch raise which indicates capitals? This is currently adjustable with the -k option to allow experimentation. 7. How should multiple capitals in a word be indicated? Or a capital which is not the first character of a word? Or does that only need to be considered when speaking letters individually (spelling)? 8. Have I misunderstood the whole point of this, and punctuation and capital indications are only needed when spelling out individual characters? _______________________________________________ Speechd mailing list Speechd@lists.freebsoft.org http://lists.freebsoft.org/mailman/listinfo/speechd From garycramblitt@comcast.net Sun Apr 30 11:40:24 2006 From: Gary Cramblitt To: speechd-discuss@nongnu.org Date: Sun Aug 13 12:52:50 2006 Subject: eSpeak - added punctuation and capitals indications In-Reply-To: <4e1b823a18jsd@clara.co.uk> References: <4e1b823a18jsd@clara.co.uk> Message-ID: <200604300540.24857.garycramblitt@comcast.net> On Saturday 22 April 2006 06:14, Jonathan Duddington wrote: > eSpeak is a compact open-source text-to-speech engine which I hope to > adapt to work well with Speech Dispatcher. > > > I've added the ability to indicate punctuation and capital letters in > the text by either speaking their name, or by playing a sound. > > I'm not sure whether what I've done is exactly what's wanted, so if > anyone wants to try it and let me know of suggestions for improvements, > please do so. > > eSpeak text-to-speech is at: > http://espeak.sourceforge.net > > download the file: test-1.09d-linux.zip > > The ReadMe file inside gives the details. It can speak either all > punctuation or just a specified set of punctuation characters. It can > speak their names, or you can set up sound files (sound icons) to be > played instead. > > Capital letters can be indicated by a sound, or by the word "capital", > or by raising the pitch of the capitalized word. > > Also the feature of embedding commands within the text has been updated. > > Questions which I'm unsure about: > > 1. Should end-of-line be indicated? > > 2. What about apostrophes within words. Currently these are not > indicated when speaking text since that would disrupt the pronunciation > of the word. > > 3. The punctuation name is spoken in a slightly different tone from > the main text, to differentiate it. Is that OK? > > 4. The actual names for punctuation characters are defined in the > data/english_list file, so these can be changed if needed (then do > speak --compile). > > 5. If the text is spoken at a fast rate, should the sound icons also > be shortened in duration? > > 6. What is the best value for the pitch raise which indicates > capitals? This is currently adjustable with the -k option to allow > experimentation. > > 7. How should multiple capitals in a word be indicated? Or a capital > which is not the first character of a word? Or does that only need to > be considered when speaking letters individually (spelling)? > > 8. Have I misunderstood the whole point of this, and punctuation and > capital indications are only needed when spelling out individual > characters? -- Gary Cramblitt (aka PhantomsDad) KDE Text-to-Speech Maintainer http://accessibility.kde.org/developer/kttsd/index.php _______________________________________________ Speechd mailing list Speechd@lists.freebsoft.org http://lists.freebsoft.org/mailman/listinfo/speechd From obert01@mistigri.org Thu Jun 29 15:55:08 2006 From: Olivier BERT To: speechd-discuss@nongnu.org Date: Sun Aug 13 12:52:50 2006 Subject: a new module in speech-dispatcher In-Reply-To: <1151581646.3334.157.camel@chopin> References: <44A38571.4040804@brailcom.org> <1151581646.3334.157.camel@chopin> Message-ID: <20060629135507.GA25000@sunset.terramitica.net> On Thu, Jun 29, 2006 at 01:47:26PM +0200, Hynek Hanke wrote: > Hi Olivier, > > this is great news. I've reviewed the code, it looks very good. I have > these concerns: > > * The copyright statement doesn't contain the proper copyright > attribution (Copyright (C) ...., date). If you do not object, > I'd prefer copyright by Brailcom, o.p.s. This puts the whole project > in a better legal position against possible violators. No problem, I fixed it. > > * line 136, module_init() contains the statement 'Debug = 1' which > seems to do nothing. Delete? I just removed it. It was there to enable debug logging. I didn't know that it could be done in the configuration file of the module. > > * Do we really need cicero_wrapper? I think we could just open a logfile > in the driver and redirect the stderr of the subprocess there before we > exec it. This way, we can do without installing an additional program. > The file where logs from cicero should be stored of course would be > configurable through cicero.conf. Yes, of course, it would be better. I will fix it in a next commit. > > * line 218, module_speak() reads `char temp[500];' That sounds pretty > dangerous. > I don't know how I could write such a thing :). I fixed it. > * Do you plan to add support for voice and pitch change in the future? For the moment, Cicero doesn't support change of pitch and volume at runtime even if it seems easy to implement it. I already suggested that to the author. I hope it will be possible soon. But the author of cicero seems very busy. > * Also, would it theoretically be possible to return audio data from > cicero and play them in the module or upstream? I do not think we should > do that now as we are currently porting the modules from Dispatcher > under TTS API and this audio handling part will change. But for the > future, this approach would be better, so I'm curious. I don't know. Maybe it could be asked to Nicolas Pitre (author of cicero). > Thanks for all your work! > Hynek Thanks very much. > > PS: Please CC speechd@lists.freebsoft.org . Regards, -- Olivier BERT e-mail: obert01@mistigri.org Etudiant a l'E.P.I.T.A. (cycle ingenieur, 3eme annee) Tel: 06 07 69 79 71 _______________________________________________ Speechd mailing list Speechd@lists.freebsoft.org http://lists.freebsoft.org/mailman/listinfo/speechd From cerha@brailcom.org Thu Jun 29 17:00:50 2006 From: Tomas Cerha To: speechd-discuss@nongnu.org Date: Sun Aug 13 12:52:50 2006 Subject: Speech Dispatcher backend for Orca Message-ID: <44A3EB22.9080304@brailcom.org> Hello, I would like to announce availability of an EXPERIMENTAL Speech Dispatcher backend for Orca. Please, see http://www.freebsoft.org/~cerha/orca/speech-dispatcher-backend.html for more information. Any feedback is welcome, however, please note that I will not be on-line until July 10. Kindest regards, Tomas -- Brailcom, o.p.s. http://www.brailcom.org Free(b)soft project http://www.freebsoft.org Eurochance project http://eurochance.brailcom.org _______________________________________________ Speechd mailing list Speechd@lists.freebsoft.org http://lists.freebsoft.org/mailman/listinfo/speechd From hanke@brailcom.org Sun Apr 30 21:16:14 2006 From: Hynek Hanke To: speechd-discuss@nongnu.org Date: Sun Aug 13 12:52:50 2006 Subject: eSpeak - added punctuation and capitals indications In-Reply-To: <4e1b823a18jsd@clara.co.uk> References: <4e1b823a18jsd@clara.co.uk> Message-ID: <1146424574.4184.83.camel@chopin> > I've added the ability to indicate punctuation and capital letters in > the text by either speaking their name, or by playing a sound. Hello, I've finally tested it and here are my comments. 1) Punctuation at the end of message is not spoken espeak --punct "Hello, how are you?" The question mark is not spoken. If I put an additional space at the end espeak --punct "Hello, how are you? " I hear "question" correctly at the end. 2) When I use signalization of capital letters with a word espeak -k 2 "Hello, how are you? I'm fine." I hear both the sound icon (as in -k 1) and the word "capital" or "all capital". I should be hearing only the words, not the sound. 3) The pitch signalization of capital letters seems to both higher the pitch of the word containing the capital letter (good) and lower the pitch of the rest (bad). It should only work as signalization of capital letters, so the rest of the text not containing capital letters should be left intact. These are only small corrections. I didn't find any more problems in the new punctuation or capital letters capabilities nor in using them together. I've also tested sound icons for punctuation characters and it seems to work good. You might want to mention the sound-icons package you use in the config file in the documentation, so that users know where to get them. We however do not have a project page for sound-icons for some reason on www.freebsoft.org. I believe it is a bug and will be fixed soon. You can give the direct download address http://www.freebsoft.org/pub/projects/sound-icons/ in the meantime. > 3. The punctuation name is spoken in a slightly different tone from > the main text, to differentiate it. Is that OK? I like this. > 5. If the text is spoken at a fast rate, should the sound icons also > be shortened in duration? I don't know. > 7. How should multiple capitals in a word be indicated? It seems to me the current version 1.10 does it reasonably well. I think finer suggestions can only be provided by real users. > 8. Have I misunderstood the whole point of this [...] No :) As we previously discussed and as was suggested by Gary Cramblitt, the main thing missing now is an API other than the command line. Thank you, Hynek Hanke _______________________________________________ Speechd mailing list Speechd@lists.freebsoft.org http://lists.freebsoft.org/mailman/listinfo/speechd From garycramblitt@comcast.net Sun May 7 22:38:34 2006 From: Gary Cramblitt To: speechd-discuss@nongnu.org Date: Sun Aug 13 12:52:50 2006 Subject: Default Volume Message-ID: <200605071638.34980.garycramblitt@comcast.net> According to the SSIP manual, the default volume is 0 (range -100 to 100). I don't think that is correct. When I run spd-say, the default volume is 100. -- Gary Cramblitt (aka PhantomsDad) _______________________________________________ Speechd mailing list Speechd@lists.freebsoft.org http://lists.freebsoft.org/mailman/listinfo/speechd From garycramblitt@comcast.net Sat Apr 1 16:17:10 2006 From: Gary Cramblitt To: speechd-discuss@nongnu.org Date: Sun Aug 13 12:52:50 2006 Subject: Documentation In-Reply-To: <442E27FF.5050704@brailcom.org> References: <200603292127.03239.garycramblitt@comcast.net> <200603311644.58103.garycramblitt@comcast.net> <442E27FF.5050704@brailcom.org> Message-ID: <200604010917.10219.garycramblitt@comcast.net> On Saturday 01 April 2006 02:13, Tomas Cerha wrote: > Gary Cramblitt wrote: > > I looked for an eps2pdf program but could only find them for Windows and > > I'm running Debian Linux. Could you convert the file for me and send it? > > It might work better for me. > > Hi Gary, > > I committed the pdf file into the CVS. The program epstopdf is a part > of the tetex-bin Debian package. > > > I can use ImageMagick to convert an eps file to pdf, but when I add 155mm > > to the texi source as you suggest, it is ignored. The result is only > > part of the image fits on the page and texinfo mangles the document > > pretty badly. According to my texinfo manual, Version 3.141592-1.21a-2.2 > > (Web2C 7.5.4): > > > > "Image scaling is presently implemented only in TeX, not in HTML or any > > other sort of output." > > This should be ok, I guess, since the PDF manual is created using TeX. > The HTML output uses the PNG image, which is scaled. The info output > uses the ascii version. > > I'm using the Debian texinfo package version 4.8-4. I'd try running > texinfo with the PDF figure from CVS. If that doesn't help, the problem > might be in your version of texinfo. > > Hope that helps. Yes! Image is properly sized and crystal clear now. It would seem that epstopdf produces a pdf that can be scaled by texinfo, while ImageMagick does not (or something). Anyway, its beautiful now. BTW, I have texinfo 4.8-8. Thank you for your help. -- Gary Cramblitt (aka PhantomsDad) _______________________________________________ Speechd mailing list Speechd@lists.freebsoft.org http://lists.freebsoft.org/mailman/listinfo/speechd From jsd@clara.co.uk Tue May 2 15:17:43 2006 From: Jonathan Duddington To: speechd-discuss@nongnu.org Date: Sun Aug 13 12:52:50 2006 Subject: eSpeak - added punctuation and capitals indications In-Reply-To: <1146424574.4184.83.camel@chopin> References: <4e1b823a18jsd@clara.co.uk> <1146424574.4184.83.camel@chopin> Message-ID: <4e20b956cdjsd@clara.co.uk> In article <1146424574.4184.83.camel@chopin>, Hynek Hanke wrote: > 1) Punctuation at the end of message is not spoken > espeak --punct "Hello, how are you?" > The question mark is not spoken. If I put an additional space > at the end > espeak --punct "Hello, how are you? " > I hear "question" correctly at the end. Yes, that's a bug. Thanks. > 2) When I use signalization of capital letters with a word > espeak -k 2 "Hello, how are you? I'm fine." > I hear both the sound icon (as in -k 1) and the word > "capital" or "all capital". I should be hearing only the words, > not the sound. That was deliberate. Ideally, "capital" would be spoken in the altered voice in the same way that punctuation is, but that's difficult to achieve at the point where capital letters are detected, so I thought the sound icon might be useful to distinguish from an actual word "capital" in the text. I'll take another look to see if I can do the altered voice instead. > 3) The pitch signalization of capital letters seems to both > higher the pitch of the word containing the capital letter > (good) and lower the pitch of the rest (bad). It should only > work as signalization of capital letters, so the rest of the > text not containing capital letters should be left intact. That shouldn't happen, and I found no evidence of it in a test that I did. The "praat" program from www.praat.org is useful to analyse speech, including a pitch trace. > You might want to mention the sound-icons package you use in the > config file in the documentation, so that users know where to get > them. We however do not have a project page for sound-icons for > some reason on www.freebsoft.org. I believe it is a bug and will be > fixed soon. You can give the direct download address > http://www.freebsoft.org/pub/projects/sound-icons/ > in the meantime. I don't know anything about the use sound icons or sound icon packages, although I noticed there was a directory /usr/share/sound/sound-icons on my computer. Perhaps recommended packages would be better documented at the Speech Dispatcher level. -- _______________________________________________ Speechd mailing list Speechd@lists.freebsoft.org http://lists.freebsoft.org/mailman/listinfo/speechd From garycramblitt@comcast.net Sun Apr 30 14:41:19 2006 From: Gary Cramblitt To: speechd-discuss@nongnu.org Date: Sun Aug 13 12:52:50 2006 Subject: eSpeak - added punctuation and capitals indications In-Reply-To: <4e1facd663jsd@clara.co.uk> References: <4e1b823a18jsd@clara.co.uk> <200604300724.23083.garycramblitt@comcast.net> <4e1facd663jsd@clara.co.uk> Message-ID: <200604300841.19356.garycramblitt@comcast.net> On Sunday 30 April 2006 08:24, Jonathan Duddington wrote: > Yes. I intend to produce a version of eSpeak as a library, with direct > function calls from a Speech Dispatcher driver. It would not include > the portaudio interface, but would return synthesized speech in memory > buffers, through a function callback. Good news! > The embedded commands (and the UTF-8 capability) are a preparation for > this. The SSML tags need to be broken down into the lower level > commands to change pitch, rate, etc. I had thought that Speech > Dispatcher needs to do this for synthesizers which don't process SSML > themselves, and that I might also be able to take advantage of S.D. > ability to this :-) But if you think it's better for eSpeak to process > SSML directly, then that's OK, subject to a couple of problems which > I'll address with the interface proposal. We plan to try to provide SSML emulation for those synths that do not support it. The problem is that SSML emulation would lead to delays before speech begins. Consider a really long message (say SSML from a web page). The goal is for SD to do as little message parsing as possible. Ideally, none. So if you can manage to support SSML, that would be welcome, but I understand it is not easy. > I hope to produce a library version of eSpeak soon, with basic features > (i.e. those equivalent to the current command-line version), so it > would be better to wait for that rather than doing any work on a S.D. > driver for the command-line version. Since we expect to start refactoring SD soon, I would not expect an espeak driver for a while, so there is no rush. :) -- Gary Cramblitt (aka PhantomsDad) _______________________________________________ Speechd mailing list Speechd@lists.freebsoft.org http://lists.freebsoft.org/mailman/listinfo/speechd From pparent@us.ibm.com Fri Jun 30 18:22:51 2006 From: Peter Parente To: speechd-discuss@nongnu.org Date: Sun Aug 13 12:52:50 2006 Subject: Speech Dispatcher backend Message-ID: Skipped content of type multipart/alternative-------------- next part -------------- _______________________________________________ Speechd mailing list Speechd@lists.freebsoft.org http://lists.freebsoft.org/mailman/listinfo/speechd From garycramblitt@comcast.net Fri Apr 14 23:31:49 2006 From: Gary Cramblitt To: speechd-discuss@nongnu.org Date: Sun Aug 13 12:52:50 2006 Subject: KEY commands containing unspeakable characters In-Reply-To: <200604141712.11798.garycramblitt@comcast.net> References: <200604141712.11798.garycramblitt@comcast.net> Message-ID: <200604141731.50010.garycramblitt@comcast.net> BTW, I created a test script, keys.test, that sends every possible KEY command. Useful for testing. -- Gary Cramblitt (aka PhantomsDad) _______________________________________________ Speechd mailing list Speechd@lists.freebsoft.org http://lists.freebsoft.org/mailman/listinfo/speechd From hanke@brailcom.org Thu Jun 29 13:47:26 2006 From: Hynek Hanke To: speechd-discuss@nongnu.org Date: Sun Aug 13 12:52:50 2006 Subject: a new module in speech-dispatcher In-Reply-To: <44A38571.4040804@brailcom.org> References: <44A38571.4040804@brailcom.org> Message-ID: <1151581646.3334.157.camel@chopin> > Yes I think it is stable enough to be included in a release. > I've been testing it for a month using orca and the gnome desktop, and > there is no problem with the speech dispatcher module itself. I have had > some problems with gnome-speech that I think is not very stable. > A few french testers reported me that I made a good job :) Hi Olivier, this is great news. I've reviewed the code, it looks very good. I have these concerns: * The copyright statement doesn't contain the proper copyright attribution (Copyright (C) ...., date). If you do not object, I'd prefer copyright by Brailcom, o.p.s. This puts the whole project in a better legal position against possible violators. * line 136, module_init() contains the statement 'Debug = 1' which seems to do nothing. Delete? * Do we really need cicero_wrapper? I think we could just open a logfile in the driver and redirect the stderr of the subprocess there before we exec it. This way, we can do without installing an additional program. The file where logs from cicero should be stored of course would be configurable through cicero.conf. * line 218, module_speak() reads `char temp[500];' That sounds pretty dangerous. * Do you plan to add support for voice and pitch change in the future? * Also, would it theoretically be possible to return audio data from cicero and play them in the module or upstream? I do not think we should do that now as we are currently porting the modules from Dispatcher under TTS API and this audio handling part will change. But for the future, this approach would be better, so I'm curious. Thanks for all your work! Hynek PS: Please CC speechd@lists.freebsoft.org . _______________________________________________ Speechd mailing list Speechd@lists.freebsoft.org http://lists.freebsoft.org/mailman/listinfo/speechd From hanke@brailcom.org Sat May 13 18:48:19 2006 From: Hynek Hanke To: speechd-discuss@nongnu.org Date: Sun Aug 13 12:52:50 2006 Subject: Updates to Speech Dispatcher documentation Message-ID: <1147538899.5472.18.camel@chopin> Hello, I've expanded the section Introduction/Design of Speech Dispatcher documentation with information about the plans for future design (Message Dispatcher + TTS API). I've included a new version of the architecture diagram. I've also moved most information about compiling the IBM TTS module from README.packagers to INSTALL as this information is also useful for users compiling Dispatcher themselves. I've kept a notice about it in README.packagers. With regards, Hynek Hanke _______________________________________________ Speechd mailing list Speechd@lists.freebsoft.org http://lists.freebsoft.org/mailman/listinfo/speechd From garycramblitt@comcast.net Sun Apr 30 13:24:22 2006 From: Gary Cramblitt To: speechd-discuss@nongnu.org Date: Sun Aug 13 12:52:51 2006 Subject: eSpeak - added punctuation and capitals indications In-Reply-To: <4e1b823a18jsd@clara.co.uk> References: <4e1b823a18jsd@clara.co.uk> Message-ID: <200604300724.23083.garycramblitt@comcast.net> On Saturday 22 April 2006 06:14, Jonathan Duddington wrote: > eSpeak is a compact open-source text-to-speech engine which I hope to > adapt to work well with Speech Dispatcher. > > > I've added the ability to indicate punctuation and capital letters in > the text by either speaking their name, or by playing a sound. > > I'm not sure whether what I've done is exactly what's wanted, so if > anyone wants to try it and let me know of suggestions for improvements, > please do so. > > eSpeak text-to-speech is at: > http://espeak.sourceforge.net > > download the file: test-1.09d-linux.zip I downloaded 1.10, which I see you posted just yesterday. I will try to answer some of your questions, but frankly, as a newcomer to Speech Dispatcher, I'm a bit fuzzy on the punctuation, spelling, and capitalization features myself. > > The ReadMe file inside gives the details. It can speak either all > punctuation or just a specified set of punctuation characters. It can > speak their names, or you can set up sound files (sound icons) to be > played instead. > > Capital letters can be indicated by a sound, or by the word "capital", > or by raising the pitch of the capitalized word. As you are probably aware, at the moment Speech Dispatcher runs espeak via its generic output module. There is currently no mechanism in SD for using your new -k and --punct command line options to control speaking of capitalization/punctuation or for modifying the espeak data files on the fly. Actually, a -k or --punct option can be configured in the generic-espeak.conf file, but this setting will apply to all speech; it will not change from one message to another as the application switches modes via the SET SELF PUNCTUATION SET SELF SPELLING SET SELF CAP_LET_RECOGN SSIP commands. We could probably enhance the generic module to support this better, but see below. > > Also the feature of embedding commands within the text has been updated. > > Questions which I'm unsure about: > > 1. Should end-of-line be indicated? Good question. AFAIK, festival does not currently speak an EOL, so my guess is "no", but perhaps others in this list have a better answer. > > 2. What about apostrophes within words. Currently these are not > indicated when speaking text since that would disrupt the pronunciation > of the word. That seems correct to me, but again perhaps others have a better answer for you. > > 3. The punctuation name is spoken in a slightly different tone from > the main text, to differentiate it. Is that OK? Excellent. > > 4. The actual names for punctuation characters are defined in the > data/english_list file, so these can be changed if needed (then do > speak --compile). > > 5. If the text is spoken at a fast rate, should the sound icons also > be shortened in duration? That's something I've wondered about myself. Anybody? And see my comments below. > > 6. What is the best value for the pitch raise which indicates > capitals? This is currently adjustable with the -k option to allow > experimentation. > > 7. How should multiple capitals in a word be indicated? Or a capital > which is not the first character of a word? Or does that only need to > be considered when speaking letters individually (spelling)? > > 8. Have I misunderstood the whole point of this, and punctuation and > capital indications are only needed when spelling out individual > characters? Capitalization, punctuation, and spelling modes are independent, although I myself am fuzzy on just how these modes are supposed to interact with one another when combined together. I downloaded espeak yesterday because I wanted to see what would be involved in writing an output module for it for SD. The goal would be to take full advantage of the espeak capabilities. The main problem I noticed is that you do not provide a library or api for interfacing directly with espeak. Everything is done through command line and manipulation of configuration files. There are two problems this creates: 1. The command line interface is two crude for controlling speech. 2. The espeak program must be loaded (and it must go through its initialization) for each message spoken. Fortunately, espeak is very fast and light-weight, but this isn't the most efficient mechanism. You are aware, I believe, that we are currently discussing a new TTS Engine API on accessibility@freedesktop.org. Hynek and I are hoping that this new API will be approved soon, whereupon we will begin a major refactoring of Speech Dispatcher to use this api. Therefore, it would be best if you could design an api for espeak that would be aligned with that specification. You do not have to implement that exact api, but you do need to provide as much of the functionality it requires as possible. The more you implement in your api, the less emulation we will have to layer on top, and the better the user experience. Some of the major changes I would like to see in espeak are: 1. Support for SSML. I noticed that you now support embedded commands for controlling rate, pitch, volume, etc. In order to use these, SD would have to parse the SSML itself and translate embedded SSML into your command syntax. That would be inefficient and probably imperfect. It would be better if espeak directly supported SSML. 2. Espeak needs to return audio directly to SD. Writing a .wav file to disk is inefficient. A more direct method, such as callbacks or socket io would be better. 3. Support for index marking. Ideally, espeak should provide callbacks and/or index mark information for the following: a. Begin/end of entire message. b. Begin/end of sentence. c. Begin/end of word. d. Custom index marks as tags in SSML. The index mark information needs to be synchronized with the audio. If you play audio yourself, then you would emit callbacks at the appropriate times just before or after playing the corresponding word, sentence, or message. If you provide audio directly back to SD, then you supply the index mark positions and timings along with the audio data. See the TTS Engine API for a suggested format. Again, you don't have to do it exactly as given in the TTS Engine API, but looking at the spec will tell you what we need to know. For example, you could implement the index mark events as separate callbacks. As long as those callbacks are synchronized with the audio callbacks, i.e., audio - index mark - audio - index mark, etc., we can construct the information we need. 4. Support for stop, pause, and resume from a specified index mark position. BTW, sorry about the previous reply. Accidently clicked on the Send button. :) Thank you for espeak and thanks for listening. -- Gary Cramblitt (aka PhantomsDad) _______________________________________________ Speechd mailing list Speechd@lists.freebsoft.org http://lists.freebsoft.org/mailman/listinfo/speechd From cerha@brailcom.org Thu Jun 29 09:46:57 2006 From: Tomas Cerha To: speechd-discuss@nongnu.org Date: Sun Aug 13 12:52:51 2006 Subject: [Fwd: Re: a new module in speech-dispatcher] Message-ID: <44A38571.4040804@brailcom.org> On Wed, Jun 28, 2006 at 06:06:16PM +0200, Tomas Cerha wrote: > Hi Olivier, thank you for your effort. What is the current status of > the module? Is it stable enough, so that we can consider releasing a > new Speech Dispatcher version? Hi, Yes I think it is stable enough to be included in a release. I've been testing it for a month using orca and the gnome desktop, and there is no problem with the speech dispatcher module itself. I have had some problems with gnome-speech that I think is not very stable. A few french testers reported me that I made a good job :) Best regards, -- Olivier BERT e-mail: obert01@mistigri.org Etudiant a l'E.P.I.T.A. (cycle ingenieur, 3eme annee) Tel: 06 07 69 79 71 _______________________________________________ Speechd mailing list Speechd@lists.freebsoft.org http://lists.freebsoft.org/mailman/listinfo/speechd From garycramblitt@comcast.net Sun May 7 23:44:00 2006 From: Gary Cramblitt To: speechd-discuss@nongnu.org Date: Sun Aug 13 12:52:51 2006 Subject: Default Volume In-Reply-To: <200605071737.25819.garycramblitt@comcast.net> References: <200605071638.34980.garycramblitt@comcast.net> <200605071737.25819.garycramblitt@comcast.net> Message-ID: <200605071744.01001.garycramblitt@comcast.net> On Sunday 07 May 2006 17:37, Gary Cramblitt wrote: > On Sunday 07 May 2006 16:38, Gary Cramblitt wrote: > > According to the SSIP manual, the default volume is 0 (range -100 to > > 100). I don't think that is correct. When I run spd-say, the default > > volume is 100. > > Nevermind, just saw this in speech.conf file: > > # DefaultVolume constrols the default volume of the voice. > # It is a value between -100 (softly) and +100 (loudly). > # Currently, +100 maps to the default volume of the synthesizer. > > DefaultVolume 100 I'm not sure what that last sentence really means. In the spd-say help it says @item -i or --volume Set the volume (intensity) of the speech (between -100 and +100, 0 is default). but the default is actually set by the DefaultVolume parameter in speechd.conf and is currently a factory default of 100. Shouldn't the spd-say and ssip manuals be updated to say that the defaults are set in the speechd.conf file? -- Gary Cramblitt (aka PhantomsDad) _______________________________________________ Speechd mailing list Speechd@lists.freebsoft.org http://lists.freebsoft.org/mailman/listinfo/speechd From obert01@mistigri.org Fri Jun 23 00:05:21 2006 From: Olivier BERT To: speechd-discuss@nongnu.org Date: Sun Aug 13 12:52:51 2006 Subject: a new module in speech-dispatcher Message-ID: <20060622220521.GA27784@sunset.terramitica.net> Hi, I added a new output module to speech-dispatcher. It handles the cicero TTS software. Cicero is a very good french TTS that uses mbrola for speech output. The cicero project is hosted at : http://www.cam.org/~nico/cicero I added the cicero to the speech-dispatcher CVS. Specific documentation and configuration files for the cicero module are hosted at: http://www.terramitica.net/projects/speech-dispatcher The documentation is only in french for the moment. Regards, -- Olivier BERT e-mail: obert01@mistigri.org Etudiant a l'E.P.I.T.A. (cycle ingenieur, 3eme annee) Tel: 06 07 69 79 71 _______________________________________________ Speechd mailing list Speechd@lists.freebsoft.org http://lists.freebsoft.org/mailman/listinfo/speechd From garycramblitt@comcast.net Fri May 12 01:25:18 2006 From: Gary Cramblitt To: speechd-discuss@nongnu.org Date: Sun Aug 13 12:52:51 2006 Subject: [Announce] Speech Dispatcher IBM TTS Output Module ready for testing Message-ID: <200605111925.18959.garycramblitt@comcast.net> A new output module in Speech Dispatcher is ready for user testing. The ibmtts output module uses the commercial IBM TTS synthesizer for Linux and offers the following capabilities: 1. Support for all the documented dialects and voices of IBM TTS, plus custom voices via the ibmtts.conf file. 2. Support for spelling mode. 3. Support for changing volume, pitch, rate, etc. 4. Support for Speech Synthesis Markup Language (SSML) (as far as IBM TTS supports it). 5. Support for index marking, including SSML tags. 6. Support for pronouncing all the Speech Dispatcher KEY commands (English, but can be customized via the ibmtts.conf file). 7. Support for all the characters in the Latin-1 charset (English, but can be customized via the ibmtts.conf file). 8. Support for sound icons via separate .wav files. In order to test the new ibmtts output module, you will need: 1. A copy of the commercial IBM TTS synthesizer for Linux. More information available at http://ibmtts-sdk.sourceforge.net/ 2. Ability to download and build Speech Dispatcher from CVS. More info at http://www.freebsoft.org/speechd I'm interested in feedback on any compile/build problems that users might have as well as usability. If anyone develops enhancements for the ibmtts.conf file for other languages, please send them to me. If you do not have a copy of the IBM TTS synthesizer itself, you can still test building of the ibmtts output module by downloading the IBM TTS Software Development Kit (SDK), which is available at the SourceForge link given above. You must retrieve the SDK using SourceForge CVS. See README.packagers file in the Speech Dispatcher source. Without the actual synthesizer, you won't hear anything, of course. The SDK (not the synthesizer) is licensed under a BSD-style license. In addition, the latest CVS version of Speech Dispatcher includes a few enhancements, including 1. Updated documentation. 2. An --ssml option for spd-say for speaking SSML. 3. A --pipe-mode option for spd-say that permits inserting spd-say into a Unix stdin/stdout pipe. See the spd-say manual for more information. 4. Additional test scripts in the src/tests directory. 5. Enhanced espeak-generic.conf file for use with ESpeak. The Speech Dispatcher project provides a device independent layer for speech synthesis through a simple, stable and well documented interface. It takes care of most of the tasks necessary to solve in speech enabled applications. What is a very high level GUI library to graphics, Speech Dispatcher is to speech synthesis. Speech Dispatcher provides excellent support for the Free Festival speech synthesizer, as well as Festival Lite (flite), ESpeak (via generic module), Epos, DecTalk, Apollo, llia, and (now) IBM TTS. -- Gary Cramblitt (aka PhantomsDad) _______________________________________________ Speakup mailing list Speakup@braille.uwo.ca http://speech.braille.uwo.ca/mailman/listinfo/speakup From garycramblitt@comcast.net Sun May 7 23:44:00 2006 From: Gary Cramblitt To: speechd-discuss@nongnu.org Date: Sun Aug 13 12:52:51 2006 Subject: Default Volume In-Reply-To: <200605071737.25819.garycramblitt@comcast.net> References: <200605071638.34980.garycramblitt@comcast.net> <200605071737.25819.garycramblitt@comcast.net> Message-ID: <200605071744.01001.garycramblitt@comcast.net> On Sunday 07 May 2006 17:37, Gary Cramblitt wrote: > On Sunday 07 May 2006 16:38, Gary Cramblitt wrote: > > According to the SSIP manual, the default volume is 0 (range -100 to > > 100). I don't think that is correct. When I run spd-say, the default > > volume is 100. > > Nevermind, just saw this in speech.conf file: > > # DefaultVolume constrols the default volume of the voice. > # It is a value between -100 (softly) and +100 (loudly). > # Currently, +100 maps to the default volume of the synthesizer. > > DefaultVolume 100 I'm not sure what that last sentence really means. In the spd-say help it says @item -i or --volume Set the volume (intensity) of the speech (between -100 and +100, 0 is default). but the default is actually set by the DefaultVolume parameter in speechd.conf and is currently a factory default of 100. Shouldn't the spd-say and ssip manuals be updated to say that the defaults are set in the speechd.conf file? -- Gary Cramblitt (aka PhantomsDad) _______________________________________________ Speechd mailing list Speechd@lists.freebsoft.org http://lists.freebsoft.org/mailman/listinfo/speechd From garycramblitt@comcast.net Fri Apr 14 03:07:07 2006 From: Gary Cramblitt To: speechd-discuss@nongnu.org Date: Sun Aug 13 12:52:51 2006 Subject: Empty SPEAK commands Message-ID: <200604132107.07604.garycramblitt@comcast.net> There is an issue I'd like to get opinions about. If a client application sends an empty SPEAK command, i.e., !SPEAK . it results in the buffer containing reaching the output module. The issue is, what is the best way to deal with this? 1. Detect in SD server and return error to client. 2. Detect in SD server and silently ignore. 3. Detect in output module and return error. 4. Detect in output module and ignore. Thanks -- Gary Cramblitt (aka PhantomsDad) _______________________________________________ Speechd mailing list Speechd@lists.freebsoft.org http://lists.freebsoft.org/mailman/listinfo/speechd From hanke@brailcom.org Thu Jun 29 22:03:06 2006 From: Hynek Hanke To: speechd-discuss@nongnu.org Date: Sun Aug 13 12:52:51 2006 Subject: a new module in speech-dispatcher In-Reply-To: <20060629135507.GA25000@sunset.terramitica.net> References: <44A38571.4040804@brailcom.org> <1151581646.3334.157.camel@chopin> <20060629135507.GA25000@sunset.terramitica.net> Message-ID: <1151611386.3329.14.camel@chopin> > > * Do you plan to add support for voice and pitch change in the future? > For the moment, Cicero doesn't support change of pitch and volume at > runtime even if it seems easy to implement it. > I already suggested that to the author. I hope it will be > possible soon. But the author of cicero seems very busy. Thanks. I still wonder about changes of voices. The module receives request to set particular language and then requests for switching voices inside that language (MALE1, FEMALE3, etc.). Do you know something about switching languages and voices in Cicero? This might be important. > > * Also, would it theoretically be possible to return audio data from > > cicero and play them in the module or upstream? I do not think we should > > do that now as we are currently porting the modules from Dispatcher > > under TTS API and this audio handling part will change. But for the > > future, this approach would be better, so I'm curious. > I don't know. Maybe it could be asked to Nicolas Pitre (author of > cicero). Ok, please do so and post your findings here. With regards, Hynek _______________________________________________ Speechd mailing list Speechd@lists.freebsoft.org http://lists.freebsoft.org/mailman/listinfo/speechd From garycramblitt@comcast.net Fri Apr 14 23:12:11 2006 From: Gary Cramblitt To: speechd-discuss@nongnu.org Date: Sun Aug 13 12:52:51 2006 Subject: KEY commands containing unspeakable characters Message-ID: <200604141712.11798.garycramblitt@comcast.net> When I send characters in the latin-1 charset > 128 to SD via !KEY command, example @ U+00AF !SPEAK Macron: . !KEY ? @ U+00B4 !SPEAK Acute accent: . !KEY ? !QUIT SD does not speak the characters using Festival. In the IBM TTS output module I am working on, I found that IBM TTS does a decent job of speaking these characters. I only had to fix a few of them. I did this by creating a simple configurable mapping table in ibmtts.conf and replacing the characters with speakable words when the output module receives a KEY command. I'm wondering if we should consider generalizing this capability to all output modules/synths. One issue is that the speakable names of characters are language dependent. It would be best if each synth handled this itself, but of course, most don't. So what's the right level to address this issue? SD Server or output module for each synth? -- Gary Cramblitt (aka PhantomsDad) KDE Text-to-Speech Maintainer http://accessibility.kde.org/developer/kttsd/index.php _______________________________________________ Speechd mailing list Speechd@lists.freebsoft.org http://lists.freebsoft.org/mailman/listinfo/speechd From bogus@does.not.exist.com Sun Aug 13 12:52:45 2006 From: To: speechd-discuss@nongnu.org Date: Sun Aug 13 12:52:45 2006 Subject: No subject Message-ID: ; Only localhost can connect to the Festival server port (set! server_access_list '("localhost")) ; Default voice (set! voice_default 'voice_rab_diphone) ; Give names to the other voices we have installed. (set! voice_us1 'voice_us1_mbrola) (set! voice_us2 'voice_us2_mbrola) (set! voice_us3 'voice_us3_mbrola) (set! voice_don 'voice_don_diphone) (set! voice_ked 'voice_ked_diphone) (set! voice_kal 'voice_kal_diphone) (set! voice_rab 'voice_rab_diphone) ; Make Festival use esd (aka ESoundD) for output so I can hear ; speech at the same time as I play sound files. (Parameter.set 'Audio_Method 'Audio_Command) (Parameter.set 'Audio_Command "/usr/bin/esdcat -m -r $SR $FILE") ; Set sound volume. 1.0 is max volume, lower the 1.0 until required volume ; is reached. (set! default_after_synth_hooks (list (lambda (utt) (utt.wave.rescale utt 0.45 t)))) Nothing weird there ... Regards, Ricky -- : Usual state: (e) None of the above. : rb@tertius.net.au http://tertius.net.au/~rb/ : I do not wish them [women] to have power over men, but over : themselves. -- Mary Wollstonecraft --UlVJffcvxoiEqYs2 Content-Type: text/plain; charset=us-ascii Content-Disposition: attachment; filename=debug-festival Wed Dec 24 06:32:09 2003: module_init() Wed Dec 24 06:32:09 2003: -> Festival: |(require 'speech-dispatcher) | Wed Dec 24 06:32:09 2003: <- Festival: |LP | Wed Dec 24 06:32:09 2003: <- Festival: |OK | Wed Dec 24 06:32:09 2003: -> Festival: |(Parameter.set 'Wavefiletype 'nist) | Wed Dec 24 06:32:09 2003: <- Festival: |LP | Wed Dec 24 06:32:09 2003: <- Festival: |OK | Wed Dec 24 06:32:09 2003: FestivalServerHost = localhost Wed Dec 24 06:32:09 2003: FestivalServerPort = 1314 Wed Dec 24 06:32:09 2003: FestivalMaxChunkLength = 300 Wed Dec 24 06:32:09 2003: FestivalDelimiters = .?!;,- Wed Dec 24 06:32:09 2003: Festival: creating new thread for festival_speak Wed Dec 24 06:32:09 2003: festival: speaking thread starting....... Wed Dec 24 06:32:51 2003: module_write() Wed Dec 24 06:32:51 2003: Requested data: |Ricky Buchanan wrote me.| (before recoding) Wed Dec 24 06:32:51 2003: Requested data after recoding: |Ricky Buchanan wrote me.| Wed Dec 24 06:32:51 2003: Requested after stripping punct: |Ricky Buchanan wrote me.| Wed Dec 24 06:32:51 2003: -> Festival: |(speechd-set-language "en") | Wed Dec 24 06:32:52 2003: <- Festival: |LP | Wed Dec 24 06:32:52 2003: <- Festival: |OK | Wed Dec 24 06:32:52 2003: -> Festival: |(speechd-set-voice "male1") | Wed Dec 24 06:32:52 2003: <- Festival: |LP | Wed Dec 24 06:32:52 2003: <- Festival: |OK | Wed Dec 24 06:32:52 2003: Filling new sample wave Wed Dec 24 06:32:52 2003: -> Festival: |(speechd-set-rate 0) | Wed Dec 24 06:32:52 2003: <- Festival: |LP | Wed Dec 24 06:32:52 2003: <- Festival: |OK | Wed Dec 24 06:32:52 2003: -> Festival: |(speechd-set-pitch 0) | Wed Dec 24 06:32:52 2003: <- Festival: |LP | Wed Dec 24 06:32:52 2003: <- Festival: |OK | Wed Dec 24 06:32:52 2003: -> Festival: |(speechd-set-punctuation-mode 'some) | Wed Dec 24 06:32:52 2003: <- Festival: |LP | Wed Dec 24 06:32:52 2003: <- Festival: |OK | Wed Dec 24 06:32:52 2003: -> Festival: |(speechd-set-capital-character-recognition-mode nil) | Wed Dec 24 06:32:52 2003: <- Festival: |LP | Wed Dec 24 06:32:52 2003: <- Festival: |OK | Wed Dec 24 06:32:52 2003: Festival: leaving write() normaly Wed Dec 24 06:32:52 2003: Semaphore on Wed Dec 24 06:32:52 2003: Entering parent process, closing pipes Wed Dec 24 06:32:52 2003: Looping... Wed Dec 24 06:32:52 2003: returned 24 bytes Wed Dec 24 06:32:52 2003: Synthesizing: |Ricky Buchanan wrote me.| Wed Dec 24 06:32:52 2003: (festival_speak_command): Ricky Buchanan wrote me. Wed Dec 24 06:32:52 2003: -> Festival: |(speechd-speak "the text isn't displayed")| Wed Dec 24 06:32:52 2003: s-returned 0 Wed Dec 24 06:32:52 2003: Looping... Wed Dec 24 06:32:52 2003: returned -1 bytes Wed Dec 24 06:32:52 2003: Retrieving data Wed Dec 24 06:32:52 2003: Starting child... Wed Dec 24 06:32:52 2003: Entering child loop Wed Dec 24 06:32:52 2003: child: Checking for header. Wed Dec 24 06:32:54 2003: <- Festival: WV Wed Dec 24 06:32:55 2003: <- Festival: LP Wed Dec 24 06:32:55 2003: <- Festival: OK Wed Dec 24 06:32:55 2003: ss-returned 0 Wed Dec 24 06:32:55 2003: Ok, next step... Wed Dec 24 06:32:55 2003: Sending request for synthesis Wed Dec 24 06:32:55 2003: Sending buf to child in wav: 60052 bytes Wed Dec 24 06:32:55 2003: going to write 10 bytes Wed Dec 24 06:32:55 2003: written 10 bytes Wed Dec 24 06:32:55 2003: going to write 60052 bytes Wed Dec 24 06:32:55 2003: child: Got 4096 bytes Wed Dec 24 06:32:55 2003: child: Got 4096 bytes Wed Dec 24 06:32:55 2003: child: Got 4096 bytes Wed Dec 24 06:32:55 2003: child: Got 4096 bytes Wed Dec 24 06:32:55 2003: child: Got 4096 bytes Wed Dec 24 06:32:55 2003: child: Got 4096 bytes Wed Dec 24 06:32:55 2003: child: Got 4096 bytes Wed Dec 24 06:32:55 2003: child: Got 4096 bytes Wed Dec 24 06:32:55 2003: child: Got 4096 bytes Wed Dec 24 06:32:55 2003: child: Got 4096 bytes Wed Dec 24 06:32:55 2003: child: Got 4096 bytes Wed Dec 24 06:32:55 2003: child: Got 4096 bytes Wed Dec 24 06:32:55 2003: child: Got 4096 bytes Wed Dec 24 06:32:55 2003: child: Got 4096 bytes Wed Dec 24 06:32:55 2003: child: Got 2708 bytes Wed Dec 24 06:32:55 2003: written 60052 bytes Wed Dec 24 06:32:55 2003: parent: Sent 60052 bytes Wed Dec 24 06:32:55 2003: going to write 24 bytes Wed Dec 24 06:32:55 2003: child: Got 24 bytes Wed Dec 24 06:32:55 2003: End of data stream caught Wed Dec 24 06:32:55 2003: child: End of data caught Wed Dec 24 06:32:55 2003: child: Opening audio... Wed Dec 24 06:32:55 2003: child: Can't open audio device! Wed Dec 24 06:32:55 2003: written 24 bytes Wed Dec 24 06:32:55 2003: parent: Waiting for response from child... Wed Dec 24 06:32:55 2003: parent: Read bytes 0, child stopped Wed Dec 24 06:32:55 2003: End of data in parent, closing pipes [1:-1] Wed Dec 24 06:32:55 2003: Waiting for child... Wed Dec 24 06:32:55 2003: child terminated -: status:1 signal?:0 signal number:0. --UlVJffcvxoiEqYs2-- From bogus@does.not.exist.com Sun Aug 13 12:52:45 2006 From: To: speechd-discuss@nongnu.org Date: Sun Aug 13 12:52:46 2006 Subject: No subject Message-ID: /* Takes the buffer, position of cursor in the buffer * and key that have been hit and tries to handle all * the speech output, tabulator completion and other * things and returns the new string. The cursor should * be moved then by the client program to the end of * this returned string. * The client should call this function and display it's * output every time there is some input to this command line. */ /* WARNING: This is still not very well implemented and so probably useless at this time. */ char* spd_command_line(int fd, char *buffer, int pos, char* c){ I have removed it from the indices. I don't even know if it's of some use. > And, last but certainly not least, is there any interest for me to do > some English fixes on the speech dispatcher documentation when I get > the time and am reading through them. There are quite a few spelling > mistakes and the sentence structure could be better in places. You are really welcome to do language corrections! Thank you for your contributions, Hynek From pdm@freebsoft.org Wed Jul 5 22:11:54 2006 From: Milan Zamazal To: speechd-discuss@nongnu.org Date: Sun Aug 13 12:52:49 2006 Subject: Announce: speechd-el 2.0 released Message-ID: <87odw3stf9.fsf@zamazal.org> speechd-el 2.0 released ======================= The Free(b)soft project is pleased to announce speechd-el 2.0. * What is speechd-el? speechd-el is an Emacs client to Speech Dispatcher and BrlTTY (and possibly other alternative output devices), providing a complex speech and Braille interface to Emacs, focused especially on (but not limited to) the blind and visually impaired users. It allows the user to work with Emacs without looking on the screen, listening to speech output produced by the speech synthesizers supported in Speech Dispatcher and watching Braille output on Braille displays supported by BrlTTY. Key speechd-el features are: - Automated reading of most Emacs messages and actions, without necessity to customize each particular Emacs function to make it speak. - Most built-in and external Emacs packages can produce speech and Braille output immediately, without any special support. - Almost no change of the standard Emacs behavior, no interference with user settings. - Highly configurable. - Full cooperation with Speech Dispatcher. - Braille output and input support. - Mechanism for adding backends for other alternative output devices. - Small code size. - Free Software, distributed under the GPL. The speechd-el package contains an Elisp libraries for communication with Speech Dispatcher and BrlTTY, allowing you to write your own Emacs speech and Braille output extensions and your own Emacs interfaces to Speech Dispatcher and BrlTTY. speechd-el design is completely different from the design of Emacspeak, another speech output package for Emacs. Though both speechd-el and Emacspeak serve a similar purpose, they differ in their approach to the problem they address. * What is new in the 2.0 version? This is a major release. The primary change is separation and modularization of the output mechanism. From the user's point of view the main result of this change is support of Braille output and input, similar to the speech output interface. So now you can use speechd-el with both Speech Dispatcher and BrlTTY. * Where to get it? You can get the distribution tarball of the released version from http://www.freebsoft.org/pub/projects/speechd-el/speechd-el-2.0.tar.gz . The home page of the project is http://www.freebsoft.org/speechd-el . _______________________________________________ Speechd mailing list Speechd@lists.freebsoft.org http://lists.freebsoft.org/mailman/listinfo/speechd From hanke@brailcom.org Mon Jul 10 12:12:11 2006 From: Hynek Hanke To: speechd-discuss@nongnu.org Date: Sun Aug 13 12:52:49 2006 Subject: Running Speech Dispatcher with ALSA? In-Reply-To: <878xn29b19.fsf@zamazal.org> References: <878xn29b19.fsf@zamazal.org> Message-ID: <1152526332.3332.6.camel@chopin> Milan Zamazal p??e v Ne 09. 07. 2006 v 15:13 +0200: > I use ALSA exclusively on my system, without OSS emulation in kernel. > I've got Speech Dispatcher 0.6 from Debian and I can't make it speak. I > use the Festival output module. > [...] > Any idea what could be wrong in my setup? You could also turn on the option Debug 1 in festival.conf and look into /var/log/speech-dispatcher/festival.log for the line ERROR: Cannot open audio device which should give a more detailed reason as reported by ALSA. With regards, Hynek Hanke _______________________________________________ Speechd mailing list Speechd@lists.freebsoft.org http://lists.freebsoft.org/mailman/listinfo/speechd From pparent@us.ibm.com Fri Jul 7 02:54:34 2006 From: Peter Parente To: speechd-discuss@nongnu.org Date: Sun Aug 13 12:52:49 2006 Subject: Speech Dispatcher backend In-Reply-To: <1152110582.4301.19.camel@chopin> Message-ID: Skipped content of type multipart/alternative-------------- next part -------------- _______________________________________________ Speechd mailing list Speechd@lists.freebsoft.org http://lists.freebsoft.org/mailman/listinfo/speechd From hanke@brailcom.org Tue Jul 4 00:00:43 2006 From: Hynek Hanke To: speechd-discuss@nongnu.org Date: Sun Aug 13 12:52:49 2006 Subject: Speech Dispatcher backend In-Reply-To: References: Message-ID: <1151964043.3361.39.camel@chopin> Peter Parente p??e v P? 30. 06. 2006 v 12:22 -0400: > I noticed you created a Speech Dispatcher backend for Orca. I'd like > to do the same for IBM's Linux Screen Reader. Hi Peter, thanks for your interest. > Do you know if the support for sound icons in Speech Dispatcher is > currently implemented? Yes, what is described in the documentation on http://www.freebsoft.org/doc/speechd/ssip_8.html#SEC8 is fully implemented. However... > LSR does currently support audio icons and can play them using a > different mixer/output library. I was wondering if I would be able to > test the same functionality with Speech Dispatcher. ...we take a little bit different approach. In Speech Dispatcher, sound icons are viewed as events and they are passed to the output driver and finally the synthesis engine as symbolic names. The engine decides (ideally based on user configuration) what to do with the event. In a situation when one user might like playing a .wav file, another user might prefer to have some text spoken, or speak text and then play a wav file etc. This is currently fully supported only in Festival and IBM TTS, other engines will at leat say the icon name in voice. As for audio output, we try to get audio samples back from the synthesizer and play them using our own modules (currently OSS, Alsa, NAS), but again, this is not possible for all synthesizers. Also if you are looking into the future, our plan with TTS API (see http://www.freebsoft.org/tts-api ) is to provide emulation for such capabilities in the above layer so that if a synthesizer won't support sound icons, a helper module can emulate them to some degree (for any synthesizer). Next release of Speech Dispatcher is going to use TTS API as its backend. Actually TTS API Provider can partly be seen as a way to refactor Speech Dispatcher into two components. I hope this helps. With regards, Hynek Hanke _______________________________________________ Speechd mailing list Speechd@lists.freebsoft.org http://lists.freebsoft.org/mailman/listinfo/speechd From garycramblitt@comcast.net Wed Jul 12 14:49:18 2006 From: Gary Cramblitt To: speechd-discuss@nongnu.org Date: Sun Aug 13 12:52:49 2006 Subject: eSpeak: API for shared library version In-Reply-To: <4e44b149edjsd@clara.co.uk> References: <4e44b149edjsd@clara.co.uk> Message-ID: <200607120849.18318.garycramblitt@comcast.net> On Tuesday 11 July 2006 05:33, Jonathan Duddington wrote: > Here is a draft API for a shared library version of the eSpeak TTS > synthesizer, which is intended for use by Speech Dispatcher. > > http://espeak.sourceforge.net/speak_lib.h > > Implementation of this is mostly done, subject to suggested amendments > and testing. > > The following SSML tags are currently implemented: > xml:lang xml:base > xml:lang, name, gender, age, variant > rate, volume, pitch, range (relative) > glyphs > > xml:lang >

xml:lang > >

xml:lang