How to evaluate an observation sequence with hmm functions

I have trained my HMM model with a set of different observation sequences with hmmtrain. Now I have the transition matrix and the emission matrix, and my initial probabilities vector as well.

But I can't find a way to evaluate the probability of a new observation sequence with the model I just trained.

Please, somebody can tell me how obtain the probability that the given sequence occurs, given the model?

Thanks!

Answer 1

[engstudent]

hi i am working in speech recognition using hmm and i need to know what is the step to build hmm andhow to train hmm in matlab7.9 ??????? i found in matlab hmmestimat,i have seq but i dont know what is the states? my Q is if i need to generat the states by my self????or this is wrong plz answer my question............

plz help me .....................

[Altin Shala]

Dear,

I am Altin Shala student from Kosova, I finished all exams in thesis and now I will prepare diploma with title "Speech Recognition in language Albanian in aplicattion Skype or Wiber" .

My request is! During exploring material on Interent I saw some project yours about Speech Recongnation you will help me with thesis . I need to simulation with Matlab some word or text in Albanin language and obtain statistics on intelligibility of speech processing Other chapter I have to study Hidden Markov model

If you have books about this thesis and code in Matlab please sent me on email.

All the best!

[Altin Shala]

My email is [e-mail]

Answer 2

[Peter Benko]

Hello!

I have the same problem. I have succesfully generated my Transition, Emission Matrices as well as the Initial probabilities from my data series. According to the literature I should be able to evaluate other data series agains this trained model, but I cannot figure out how can I do that with MATLAB. Is there any solution for it in the 2013a version?