New Developer working on MS SAPI
hanke at brailcom.org
Fri Nov 7 16:17:45 CET 2008
Thomas Lloyd wrote:
thanks for your work and interest.
> Currently I have an executable that speaks using SAPI using wine and
> Microsoft speech engine in Ubuntu. I have also installed an additional
> speech engines from NeoSpeech VWKate and once selected it will use any
> installed speech engine.
This sounds good.
> Main Tasks:
> 1. Speech Dispatcher driver - Hope to get help/pointers from you guys
I'd recommend you start with setting up a simple ''generic''
Speech Dispatcher driver with your Ubuntu executable.
This should provide you with basic speech through
dispatcher and allow for some testing.
See the Dispatcher documentation on this
and config/modules/espeak-generic.conf for an example.
Then of course it would be nice to create a proper
driver (output module) for Speech Dispatcher.
I think that from the existing drivers, closest to your
intended one would be the Festival driver (it communicates
with Festival as with a server through sockets, which is
probably similar to what you want to do).
Another big decision point is if you can get the synthesized
audio back from the SAPI (and play it with the standard
dispatcher sound library, which is the prefered way)
or you let the Windows part produce sound.
Any method you chose, please see the Speech Dispatcher
documentation section Server Programming/Output modules
> 10. Come up with a way to configure Wine sound in an accessible way
This might not be necessary, as explained above.
More information about the Speechd