В Птн, 21/12/2007 в 14:13 -0200, Marcelo Yassunori Matuda пишет:
> Hi,
> 
> I wrote an ***experimental*** C++ program that uses the Tube
> Resonance Model and the file diphones.mxml (articulation data) from
> Gnuspeech.
> 
> Apart from the tube model, it is not a translation of the Gnuspeech
> ObjC code, so the quality of speech output is not the same.
> 
> It may be used as a voice in Festival. Festival is needed if you want
> to use the program as a TTS system.
> 
> You may download it at: http://sourceforge.net/projects/dvagar
> 
> This is for us mere mortals who can't use Gnuspeech... :)

Hi Marcelo

Amazing work! I think it's a real future of gnuspeech with all festival 
framework I hope it will get great progress.

A few comments:

1. Please check for xerces-c in configure and add it as a dependency
2. Could you provide a sample of synthesis, like this one I've just got
with your program:

http://festlang.berlios.de/docu/lib/exe/fetch.php?id=gnuspeech&cache=cache&media=examples:gnuspeech-sample.wav

3. What about announcement on festival-talk?

It would be nice to get this announced on gnuspeech website and
integrated into festival as well. I know many of us wrote compatibilty
code like this to make gnuspeech at least work on linux.

Attachment: signature.asc
Description: Эта часть сообщения подписана цифровой подписью

_______________________________________________
gnuspeech-contact mailing list
[email protected]
http://lists.gnu.org/mailman/listinfo/gnuspeech-contact

Reply via email to