[Asterisk-Dev] voice recognition

[Zac Wolfe]

I've played a bit with Sphinx-4 (http://cmusphinx.sourceforge.net/sphinx4/)
and it looks very promising.  It's a complete rewrite of the original Sphinx
application in Java.  Benchmarks show it to be substantially faster than the
old C-based version and it's pretty easy to use.  I tried using it with
Asterisk (using JAsterisk http://sourceforge.net/projects/jasterisk/ ) but
the recognition accuracy over the phone was low since there were no
8kHz-based acoustic models available then.  The latest version is supposed
to include a "Narrow bandwidth acoustic model" so I'll probably try it
again.

Not VR but for TTS I've found that FreeTTS
(http://sourceforge.net/projects/freetts/) is both faster than Festival and
much easier to configure and use.  I'm using it for several production IVR's
I have set up using JAsterisk.

-----Original Message-----
From: Holger Schurig [mailto:hs4233 at mail.mn-solutions.de]
Sent: Wednesday, September 29, 2004 11:33 PM
To: asterisk-dev at lists.digium.com
Subject: Re: [Asterisk-Dev] voice recognition


http://htk.eng.cam.ac.uk/develop/atk.shtml
http://www.isip.msstate.edu/projects/speech/software/
http://mi.eng.cam.ac.uk/~sjy/software.htm

_______________________________________________
Asterisk-Dev mailing list
Asterisk-Dev at lists.digium.com
http://lists.digium.com/mailman/listinfo/asterisk-dev
To UNSUBSCRIBE or update options visit:
   http://lists.digium.com/mailman/listinfo/asterisk-dev