I just ran Julius, with Voxforge's model, on 30 different audio files, of people saying short simple phrases. I expected Julius to miss a few, but it failed to transcribe *any* of them. That's a bit disappointing.

--- (Edited on 10/1/2011 9:55 am [GMT-0500] by Cerin) ---

I just realized my jconf was using controlapp's toy grammar. How do I get Julius/Julian to support arbitrarily large grammars?

--- (Edited on 10/1/2011 10:06 am [GMT-0500] by Cerin) ---

Since the reason for opening the thread seems to have been resolved, perhaps it is a bad idea to fork into a different issue in the same thread. It's probably in your best interest to open up a different thread with an explicit subject line, that way it will be read by others who will be able to address the issue.

--- (Edited on 10/1/2011 10:44 am [GMT-0500] by colbec) ---

try wrapping the command with padsp, so instead of:

julius -quiet -input mic -C julian.jconf 2>/dev/null | ./command.py  

run:

padsp julius -quiet -input mic -C julian.jconf 2>/dev/null | ./command.py  

--- (Edited on 12/3/2012 12:34 pm [GMT-0600] by Visitor) ---

hey,, 
the julius package isnt respoding to voice properly,
is there a guide explining the way to proceed with the installation of julius and its implementation in creating a voice controlled computer?
would appreciate any help!!  

--- (Edited on 3/14/2013 3:00 am [GMT-0500] by Visitor) ---

>guide explining the way to proceed with the installation of julius and its

>implementation in creating a voice controlled computer?

http://iij.dl.sourceforge.jp/julius/47534/Juliusbook-4.1.5.pdf

Tutorial: Create Acoustic Model - Manually

--- (Edited on 3/14/2013 7:55 am [GMT-0400] by kmaclean) ---