i m doing a project on speech recognition. 
1) i've tried to determine the end-points of connected words and get the
separate words. but the prob is that it is not removing complete silence
portion. And in some cases it is considering even the noise portion as a
seperate word. 
2) What if we dont separate the words and take MFCCs of both words together as
one matrix. will that be ok?
Cn any1 help.

______________________________
New Code Sharing Section now Live on DSPRelated.com. Learn about the Reward Program for Contributors here.



(You need to be a member of speech-recognition -- send a blank email to speech-recognition-subscribe@yahoogroups.com )

Hey!

You haven't mentioned which approach does your speech recognizer take; if its
DTW, NN or HMM.

If its HMM, you can model the noisy portions as a state (in your word lattice if
you have one, or just fit stick it in appropriately). I'm sure you can
incorporate the same in the other approaches as well.

However, silence/noisy model estimation is unreliable (as the noise environment
might keep changing : babble noise, cocktail party env.) so you would want to
use a noise canceller as a preprocessor.

Again, if you are working with a clean speech environment, forget about all
these and just introduce a silence model :)

Regards,
Abhilash.

______________________________
StarterWare provides C-based no-OS platform support for the C6000 DSP devices.
Click here for details.



(You need to be a member of speech-recognition -- send a blank email to speech-recognition-subscribe@yahoogroups.com )