Common interfaces for speech services

Michael Pozhidaev msp at altlinux.ru
Mon Nov 7 17:26:18 PST 2011


Hello, Jeremy!

Yes, I had to look through freebsoft documents during the discussions of
ideas to merge speech-dispatcher with our russian server voiceman.

In its current view it has some disadvantages and it is too large with
lot of detailes. It is basically designed for using TTS engines as
shared libraries and requires additional work to make it appropriate for
D-Bus. 

My suggestion is to start with minimal simple interface and extend it
later when new detailes become clear and appropriate for everybody.

> Hello Michael,
>
> It seems there was some work done in this regard to define a tts api
> that providers could conform to.  I don't recall if it was dbus based,
> but more information can be found here:
> http://devel.freebsoft.org/tts-api  There is even a python provider
> implemtation http://devel.freebsoft.org/tts-api-provider but I'm not
> sure how fully developed it is.  At the bottom of the api page it says
> discussion for the api takes/took place on
> accessibility at freedesktop.org mailinglist.  That may be a place to
> look for more background too.
>
> thanks,
> Jeremy
>
> On Mon, Nov 7, 2011 at 4:11 PM, Michael Pozhidaev <msp at altlinux.ru> wrote:
>> Hello, everybody!
>>
>> I have the question about developing common interfaces for various
>> services. In Russia there is speech server used by blind people for
>> real-time generation text-to-speech. It takes commands like "say" and
>> "stop". Now this server uses its own protocol for client connections but
>> we have an idea to add D-Bus implementation to unify access to service.
>>
>> It seems to me it is a good approach to suggest common interface to
>> text-to-speech servers in org.freedesktop like we already have
>> org.freedesktop.MediaPlayer for every media player as any other common
>> interface for other type of services.
>>
>> As I got by searching in Google there is no common interfaces in
>> org.freedesktop for speech servers but there are some attempts to
>> process speech commands with D-Bus without any compatibility between
>> each other.
>>
>> I would like to suggest some basic view of the interface for speech
>> servers but I don't know what I must do for this work. I am ready to do
>> it, provide some basic implementation, listen different comments and
>> include improvements.
>>
>> Is any registration on freedesktop needed for this work or I can simply
>> present my suggestions here and follow the discussion?
>>
>> If I am wrong and there are any already existing interfaces, please let
>> me know. In this case I will simply add corresponding implementation.
>>
>> Thanks for any information!
>>
>> --
>> Michael Pozhidaev. Tomsk, Russia.
>> Russian info page: http://www.marigostra.ru/
>> _______________________________________________
>> xdg mailing list
>> xdg at lists.freedesktop.org
>> http://lists.freedesktop.org/mailman/listinfo/xdg
>>

-- 
Michael Pozhidaev. Tomsk, Russia.
Russian info page: http://www.marigostra.ru/


More information about the xdg mailing list