speechd-discuss
[Top][All Lists]
Advanced

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [orca-list] making Linux a hospitable place for TTS engines like Vox


From: Bill Cox
Subject: Re: [orca-list] making Linux a hospitable place for TTS engines like Voxin
Date: Sun, 10 Jan 2021 17:49:13 -0800

I have my meta-module working with speech-dispatcher well enough for folks to start kicking tires.  I still need to upload binaries for the backends, and update the README file to explain how to install it on various Linux distros.  The only part that has to be compiled per distro is the new meta-module I currently call speechsw.

Would it be possible to include my meta-module, named speechsw.c, in the src/modules directory?

I actually like having the meta-module, since it gives me a place to add features I want, such as using libsonic to adjust pitch and speed for picotts.  I'm going to add a silence reduction algorithm as well, and maybe some more goodies down the road.

Would folks be OK with me using my meta-module as a testing ground for possible future enhancements to speech-dispathcer?  For example, we could add a new sound setting which enables sound samples to be returned, exactly as Samuel suggested, assuming folks find the additional latency introduced in my meta-module acceptable.  Similarly, features like silence reduction could be pulled into the speech-dispatcher daemon if users would like it to be available to all modules, not just my meta-module.

WDYT?

On Mon, Dec 28, 2020 at 12:38 PM Samuel Thibault <samuel.thibault@ens-lyon.org> wrote:
Gilles Casse, le mer. 23 déc. 2020 11:24:43 +0100, a ecrit:
> Also a user should be free to install locally, without superuser access,
> a text-to-speech.

That's a good point indeed, added here:
https://github.com/brailcom/speechd/issues/452

Samuel


reply via email to

[Prev in Thread] Current Thread [Next in Thread]