Design suggestion: The server just for synthesis

[Michael Pozhidaev]

Hi!

> Hmm, in your first mail you wrote:
>> > 2. It can accept connection, receive portion of text and send audio
>> > data
>> back to the client.
>
> What do you mean with client here? If the client is a screenreader this
> approach doesn't make sense.
No, not a screen reader. Screen reader uses some server to process text
data it generates (let call it audio output part), but this server must use
in my suggestion another one to synthesize speech audio data. (let call it TTS 
part).

> Some weeks ago I started a thread where I have suggested almost the same
> but got no useful response.

Some time ago I was out of Russia and didn't read mail. Probably your
message was sent during my absence, sorry.

> I called the audio-playback-part speech-dispatcher-audio-agent which
> should redrive the speech-data from the core server and play them back
> which the available audio-output.

Yes, absolutely right, it is the same approach. I am ready to discuss
what protocol can be used for this purpose, what already created code is
useful and so on. Do you have some ideas for these questions?

-- 
Michael Pozhidaev. Tomsk, Russia. E-mail: msp at altlinux.ru
Russian info page: http://www.marigostra.ru/