From: jasonmadams@gmail.com   
      
   On Aug 23, 7:42 pm, "Alessia"  wrote:   
   > Hello,   
   >   
   > I'm trying to training an hidden markov model using the forward - backward   
   > algorithm (Baum Welch) but I don't understand which input I have to give it.   
   > In particular, should I give a complete training set (states + observations)   
   > or only the observations?   
   >   
   > thanks!   
   > Alessia   
   >   
      
   You'll need to give it the observations, the state transition   
   probabilities, and the observation probabilities.  This tutorial may   
   help if you haven't already seen it:   
   http://www.cs.cmu.edu/~roni/11761-s07/Presentations/whw-HMMs-in-ASR-3.1.pdf   
      
   [ comp.ai is moderated ... your article may take a while to appear. ]   
      
   --- SoupGate-Win32 v1.05   
    * Origin: you cannot sedate... all the things you hate (1:229/2)