[mary-users] French Voice

[Ingmar Steiner]

Dear Florent,

wow, that sounds very good! I'm no native speaker of French, but to me 
the segments sound great (not surprisingly, the HMM-based synthesis is 
pretty smooth); the main drawbacks are the voice source (which is way 
beyond your control at this point), and the prosody. Many of the 
prosodic problems I believe are due to a lack of control at the metrical 
level, mainly because the French language components seem to be 
oblivious to syllables. Improving that should also fix most, if not all, 
duration problems.

Again, very impressive, and I certainly hope that you will be able to 
share your code by checking it into the fr-branch!

Best wishes,

-Ingmar

On 06.08.2011 21:40, fxavier at ircam.fr wrote:
> Dear all,
>
> Here's a sample of the brand new french voice. Of course it is not a final
> result, some improvements have to be made:
> - still have to syllabifie the phonetical transcription
> - duration models are awkward
> - some strange noises during the pauses
> Anwyay I think it's a good starting point.
>
> Thanks for everything.
>
> Florent
>
>
>
> _______________________________________________
> Mary-users mailing list
> Mary-users at dfki.de
> http://www.dfki.de/mailman/cgi-bin/listinfo/mary-users

-- 
Ingmar Steiner
Postdoctoral Researcher

LORIA Speech Group, Nancy, France
National Institute for Research in
Computer Science and Control (INRIA)