Re: [orca-list] Incorporating Acapella Voices or Infovox iVox



W dniu 2012-11-06 20:52, mattias napisaÅ(a):
ok if i remember right
it requere both client and server?

Yes. Server can run in Wine even without graphic and audio output (I had Ivona installed on my headless router with Xvfb). At now it's possible to use my patched Open Sapi server until someone will rewrite it in more native and friendly way. Client (in our case speech-dispatcher module) is relatively simple and it's possible to reuse 90% of milvona sources.

On 11/06/12 19:26, Bohdan R. Rau wrote:
W dniu 2012-11-06 17:10, Alex Midence napisaÅ(a):
Hate to be the bearer of ill tidings but, last I heard this was
simply not possible. Open Sapi is supposed to be pretty buggy.

Open Sapi is not buggy - it's simple only idea how to use Windows SAPI voices in Linux, and it's forgotten by author.
Some time ago I published this information, I'll repeat again:

1) I prepared patch to Open Sapi to work natively in Linux (in fact, only change was the synthesized wave is not played on Windows audio system, but it's returned to calling Linux process). 2) I wrote speech-dispatcher module Milvona (sorry, only for Polish language) working with Orca and Ivona SAPI voice. It's rather slow, but Ivona has some limitations and probably with other synthesizers speed would be acceptable.

Conclusion:

1) It's possible to have access to Windows voices in Linux (in fact, we need only one Windows C++ programmer - I tried to create Windows part in Python, but there was overkill - about 80 MB of libraries to run program which fits in 50 lines)

2) It's possible to use SAPI voices in Orca (in fact we need only one Linux programmer familiar with speech-dispatcher).

If you can find this Windows programmer - I can help with creating Linux part. Sorry - I have practically no experience with Windows.

ethanak

--
http://milena.polip.com/ - Pa pa, Ivonko!



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