[mary-users] New french voice

[fxavier at ircam.fr]

Dear Marc,

I really enjoyed being involved in this project!

Unfortunately, lia_phon indeed is only available for Unix systems. That
said, the core is written in pure C. According to Frederic Bechet, the way
to make it available for all the OS, is to rewrite in Java the script
"lia_tex2phon". This one calls those following scripts (which are written
in csh).

$LIA_PHON_REP/script/lia_nett
$LIA_PHON_REP/script/lia_taggreac
$LIA_PHON_REP/script/lia_phon

But those srcipts launches bin files and not .c programms. Maybe you have
an idea on how to do this, but I'm sure this won't take long to port it in
Java.

Best regards,
Florent




> Dear Florent,
>
> this is great, thanks a lot. I see that you are making two voices
> available together with the code, both under the Creative Commons
> Attribution-NoDerivatives license, and the source code under the LGPL.
>
> This is great, and I am looking forward to integrating this with MARY TTS.
>
> Of course, you could have committed the code directly to the fr-branch
> on the MARY repository that we had created for you -- this would have
> been easier for you and for us... ; but never mind, we can use it in
> this form as well.
>
> I see that you have included lia_phon for the grapheme-to-phoneme step,
> and its README says that it is licensed under the GPL and is a native
> component that seems to compile under linux. This is a suboptimal aspect
> of course, because it means that this French language package for MARY
> TTS will run only on some operating systems since it is not pure java,
> and it is incompatible with closed-source use such as for most
> commercial applications.
>
> But apart from these limitations, it seems very promising and I want to
> thank you for making available this work!
>
> Best regards,
> Marc
>
>
> On 05.10.11 11:44, fxavier at ircam.fr wrote:
>> Dear all,
>>
>> I'm happy to share with you all the works we have made to create the
>> french voice for Mary TTS. It includes the NLP components, a modified
>> version of the class "PronunciationModel.java" (due to an unfixed bug,
>> please see the README for further informations) and two HMM voices
>> created
>> (male and female).
>>
>> The french language/voices have been succefully installed in several
>> linux
>> computers.
>>
>> Please find all the sources and voices here:
>> http://dl.free.fr/n36ZO7VMt
>>
>> You can download it by clicking on "Télécharger le fichier" on the right
>> of the screen.
>>
>> I wish to thank Marc, Ingmar, Satish, Marcela, Nicolas and everybody
>> that
>> helped me achieving this goal.
>>
>>
>> Sincerly yours,
>> Florent
>> _______________________________________________
>> Mary-users mailing list
>> Mary-users at dfki.de
>> http://www.dfki.de/mailman/cgi-bin/listinfo/mary-users
>
> --
> Dr. Marc Schröder, Senior Researcher at DFKI GmbH
> Project leader for DFKI in SSPNet http://sspnet.eu
> Team Leader DFKI TTS Group http://mary.dfki.de
> Editor W3C EmotionML Working Draft http://www.w3.org/TR/emotionml/
> Portal Editor http://emotion-research.net
>
> Homepage: http://www.dfki.de/~schroed
> Email: marc.schroeder at dfki.de
> Phone: +49-681-85775-5303
> Postal address: DFKI GmbH, Campus D3_2, Stuhlsatzenhausweg 3, D-66123
> Saarbrücken, Germany
> --
> Official DFKI coordinates:
> Deutsches Forschungszentrum fuer Kuenstliche Intelligenz GmbH
> Trippstadter Strasse 122, D-67663 Kaiserslautern, Germany
> Geschaeftsfuehrung:
> Prof. Dr. Dr. h.c. mult. Wolfgang Wahlster (Vorsitzender)
> Dr. Walter Olthoff
> Vorsitzender des Aufsichtsrats: Prof. Dr. h.c. Hans A. Aukes
> Amtsgericht Kaiserslautern, HRB 2313
> _______________________________________________
> Mary-users mailing list
> Mary-users at dfki.de
> http://www.dfki.de/mailman/cgi-bin/listinfo/mary-users
>