[gst-devel] GStreamer and automatic speech recognition

[Jan Schmidt]

<quote who="David Huggins-Daines">

> I've just created a short tutorial about integrating the PocketSphinx 
> speech recognition engine with Python, GTK+ and GStreamer:
> 
> http://www.speech.cs.cmu.edu/cmusphinx/moinmoin/GStreamer
> 
> This is all quite bleeding-edge (and the default language model in 
> PocketSphinx is pretty bad so you shouldn't expect accurate 
> recognition), but I'd appreciate any feedback on the tutorial or the 
> implementation.
> 

This is thoroughly awesome. I built it without too much trouble (attaching
one small patch), unfortunately as you suggested I just seem to get
gibberish from the actual recognition. It'll be sweet when it works though
:)

J.
-- 
Jan Schmidt                                  thaytan at mad.scientist.com

<stibbons> Yeah.  The whole climax thing would make much more sense
           if I'd paid attention.
-------------- next part --------------
A non-text attachment was scrubbed...
Name: pocketsphinx.diff
Type: text/x-diff
Size: 760 bytes
Desc: not available
URL: <http://lists.freedesktop.org/archives/gstreamer-devel/attachments/20080403/0cf1c0e0/attachment.diff>