[mythtv] Voice control of MythTV

Jared Hanson jared.hanson at helixent.com
Mon Mar 24 17:29:23 EST 2003

> I don't *know* any of them myself, but some googling:
> - "voice recognition" linux
> - "speech recognition" linux
> brought up:
> http://www.isip.msstate.edu/projects/speech/
>  - Just (7-jan-2003) released a new version, seems fully functional
> http://htk.en.cam.ac.uk/
>  - Has same problem as nuance, no redists.
> Henk Poley <><

Yeah, I am aware of these two systems.  I have experience with both Sphinx
(mentioned earlier) and HTK from the open source world.  I've worked with
Philips SpeechPearl (now sold to ScanSoft) and Nuance from the commercial
side.  The commercial ones beat the open source ones hand down in terms of
features.  To be fair, however, the open source ones are mainly for research
use, and not intended for deployment.

I had looked at the ISIP stuff, but at the time it was not very well
developed.  It looks like it has progressed and is progressing quite well.
It's still very research oriented, but it looks designed well enough that it
could be an option for a replacement.  However, a LOT of work would need to
be done in that case.  I'll probably look into what would need to be done as
soon as I've wrapped up current loose ends.  Ideally, the engine used for
speech processing could be chosen at compile time and thus not dependent on
any single system.

I did contact Nuance today.  I don't know if anything will come of that, but
who knows.  Thanks for the links, and I will keep looking for possible


More information about the mythtv-dev mailing list