On Nov 4, 2009, at 8:42 PM, John Delaney wrote:
One thing I've noticed is that the syntax in the speechserverprotocol is different from the notation used by the eventlist and monet, ie:
notionalPitch Setting a base pitch
pretonicRange How much pitch movement occurs *before* reaching the tonic foot in the tone group being processed -- see last email
pretonicLift How much the pre-tonic feet are raised
tonicMovement How much pitch movement occurs in the tonic
These relate to the intonation contour as set up for Monet. You need a little understanding of the Halliday basis of the intonation
are named differently than:
speed; To do with the rate of speech
elasticity; Obsolete parameter.
intonation; Is intonation applied or not
pitchOffset; Difference in pitch from the default pitch
volume; The volume and balance have the obvious meaning
These relate to the set-up for the TextToSpeech Server You may find the Developer_TextToSpeech_Manual.pdf that I referenced in my last email helps you understand some of the Server parameters. Looking at the intonation control in Monet will help you understand the intonation contour parameters used in experiments with Monet and not really part of the Server adjustments as far as the user is concerned.
I'm not sure which variables correspond to what, so I'll take a good first stab and try to get things working for my purposes now.
On Wed, Nov 4, 2009 at 6:21 PM, John Delaney <address@hidden>
Here I was trying to implement a speech synthesis API for a graduate musical synthesis class, and now I'm getting roped into actually working on the project. I'll implement some sort of Parameter class to hold the current intonation parameters, that should be pretty simple.
Would it be possible for the synthesis engine to ramp up the intonation at the end of a sentence whenever there is a question mark? I don't think I have seen a synthesis engine, yet to do this, and it seems like such a small/easy thing to do.
Perhaps I'll revisit this when I eventually take machine learning classes.
On Wed, Nov 4, 2009 at 5:09 PM, Dalmazio Brisinda <address@hidden>
Yes, you are correct. All those server methods are yet to be implemented. Currently the server just supports speaking text with the defaults that were taken from Monet. This is certainly one area that could use some filling out, and any contribution would be more than welcome.
On 2009-11-04, at 5:54 PM, John Delaney wrote:
Thank you all for your help. I have switched to using the server method because its very easy and functional. Am I mistaken, though that many of the parameters such as pitch and intonation have not yet been implemented to the server? I am looking at the server and all the get/set methods have return zero. I suppose I will need to impliment those if this is the case.
On Wed, Nov 4, 2009 at 12:37 PM, Dalmazio Brisinda <address@hidden>
Have a look at Linked Frameworks section in the Xcode Groups & Files pane. I've found in the past that for setting up the project on a different system, I've often had to remove the custom frameworks (Tube and GnuSpeech) and then add them again, so Xcode correctly picks up the new locations -- unless they're in standard system Framework folders. If you would like additional information on Xcode, have a look at the book "Xcode Unleashed" -- there may be others.
Also, you might just want to confirm that when you open up the build options panel (CMD-i), select the Build tab, and some ways down you will see in the "Search Paths" section the option: Framework Paths, specifically, double-click and see that it's set to "$(SRCROOT)/.." with the "recursive" switch turned on. This is set for all the individual xcode projects. This expands out to show you the actual additional paths used to search and link against. $(SRCROOT) is configured on my system to point to "../.." in the General tab of the project info panel (CMD-i). Alternatively, you can just use hard paths in specifying the header/framework paths.
If you have the frameworks installed, then you might want to go to where they are (/Library/Frameworks/GnuSpeech.framework and /Library/Frameworks/Tube.framework) and rename them (say with .original appended), and then create links in /Library/Frameworks/ to the compiled frameworks in your project directory. You only really need to do this if you think you will be modifying or experimenting with the Tube and GnuSpeech frameworks. That way apps pick up the latest changes to these frameworks, instead of using the previously installed frameworks.
Also, if you have problems, try opening up the Tube and GnuSpeech xcode projects, and build them separately. Then go back and try and build the various apps etc. that depend on them. This in the event of some odd Xcode project configuration settings.
If you've made any other changes to the build settings, I suggest downloading another copy of the distribution as a backup "reference-only" copy so you can compare against it to see if your build settings for any given project match. Accidental or experimental changes to some of these settings may result in warnings or even errors during compilation/linking.
On 2009-11-04, at 12:06 PM, John Delaney wrote:
Ok, found the diagnostic info. I have the most recent build of GnuSpeech, and the applications do produce speech.
Also I have a question, I notice that in Xcode, rather than having the compiled frameworks, you have an xcode project of the framework. When I try to do that I get linking errors. How do you do include the uncompiled frameworks and then get them to compile? I've looked through all the settings and can't figure out the problem. Pardon my rookie questions.
On Wed, Nov 4, 2009 at 12:17 AM, Dalmazio Brisinda <address@hidden>
Could you try installing the latest pre-compiled beta-release version and see if the various applications generate speech? Then we can hopefully rule out any sort of system configuration issue -- at least as far as the executable apps and libs are concerned.
Also, to get the backtrace of the segmentation fault, open Console, and show the Log List. Then scroll down and you should see under the headings DIAGNOSTIC INFORMATION -> User Diagnostic Information -> <the crash file> or FILES -> ~/Library/Logs -> CrashReporter -> <the crash file>
gnuspeech-contact mailing list