[Asterisk-Users] Cepstral

[trixter http://www.0xdecafbad.com]

On Mon, 2005-07-11 at 09:19 +0100, Bob Goddard wrote:
> Compared to Rhetorical (http://www.rhetorical.com/cgi-bin/demo.cgi),
> the ATT system sounds awful.

Yes it does sound considerably better, but what do I know I have a
hearing loss.  Anyway, have you managed to integrate this with asterisk
successfully?  

If so how? use the generated output (wav presumably) and stream it via
an AGI?  Or does it have more direct asterisk connectivity?

I had written a script that would snarf sitepal TTS data, rip the SWF to
get the resulting mp3, transcode that to something more appropriate, and
cache it for future requests of the same data.  I called this script via
a macro so it was trivial to specify a voice and text to speak in a
dialplan entry.  The whole process was quite trivial but not as system
friendly as I like (too many fork and execs for my taste).  Plus becuase
the TTS engine runs on their systems there is noticable lag in actually
getting the data - part of which appears to be their system is somewhat
slow in generating the audio data, but in this instance free.

I could of course do something similar with Rhetorical, even if its
running on my own machine, but I just dont like doing it that way, it
works, but seems wrong due to many wasted cpu cycles.

-- 
Trixter http://www.0xdecafbad.com     Bret McDanel
UK +44 870 340 4605   Germany +49 801 777 555 3402
US +1 360 207 0479 or +1 516 687 5200
FreeWorldDialup: 635378
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 189 bytes
Desc: This is a digitally signed message part
Url : http://lists.digium.com/pipermail/asterisk-users/attachments/20050711/5a112f89/attachment.pgp