Hi There,

Okay, I love Julius.  It actually performs quite well for my most recent needs, but now I have a question regarding dictation.  I have looked high and low for a good speaker independant speech recognition engine, but have found little answers.  It seems that a "free" open source one does not exist simply due to the energy that must be put forth in creating decent acoustic models.  So, I looked at commercial solutions (which I am willing to entertain), but simply don't know which is best.

To put it simply:  I need a linux-based speaker-independant speech recognition engine (commercial or otherwise).  I looked at LumenVox, but their's doesn't do dictation.  I looked at WizzardSoftware but I can't tell from their website if their server based solutions has speaker independant features.  I looked at Nuance, and they seem to have acquired all the IP out there.  But they have so many products I don't know what's what. 

Can anybody give me guidance as to where I should look for a speaker independant speech recognition engine might be? 

Thank you for your time and help!

--- (Edited on 11/22/2009 10:41 pm [GMT-0600] by ariestav) ---

Hi ariestav,

>I need a linux-based speaker-independant speech recognition engine

>(commercial or otherwise). looked at LumenVox, but their's doesn't do

>dictation.

DNS on Wine

Ken

--- (Edited on 11/24/2009 12:39 pm [GMT-0500] by kmaclean) ---

Thanks Ken!  Much appreciated on that guidance.  One question:  do you know if I am able to send a digital audio file to DNS on Wine via a command line interface?  In Julius, I just type in a file name in the command line, and Julius does it's magic.  Does DNS on Wine work the same way?

Thanks again for your help!

--- (Edited on 11/24/2009 10:23 pm [GMT-0600] by ariestav) ---