Discover MakerZone

MATLAB and Simulink resources for Arduino, LEGO, and Raspberry Pi

Learn more

Discover what MATLAB® can do for your career.

Opportunities for recent engineering grads.

Apply Today

Thread Subject:
Speech Recognition via MFCC(feature extraction) and HMM classfication

Subject: Speech Recognition via MFCC(feature extraction) and HMM classfication

From: Premen

Date: 18 Sep, 2010 12:04:03

Message: 1 of 8

Hai,

currently i doing my thesis(follows as the title) i have done the MFCC as the FE. now i doing the training part and testing phase.
for training i using the Baum-Welch algorithm for training and viterbi for recognition.

1)my question is my log-likelihood from my training set is in positive value. i am not sure whether it acceptable o not? most of the value are in positive but in some papers the value was in negative.

2) i have got some viterbi path. not sure how to make use of it? help me in that too

3) Actually i am confuse how to do the the recognition part? can anyone willing explain to me. how to do a effective recognition?

My matlab code:

%%%%%%%%Training of HMM (Baum-Welch algorithm)%%%%%%
[LL, prior1, transmat1, mu1, Sigma1, mixmat1] = ...
mhmm_em(data1, pi, transmat, mu, Sigma, mixmat, 'max_iter', 5);
mixmat1 = max(mixmat1, 1.0e-5);

%%%% Calculate the Log Likelihood %%%%%%
[loglik, errors] = mhmm_logprob(data1, prior1, transmat1, mu1, Sigma1, mixmat1);
loglik

%%%%%%%%%%%%%% Viterbi Alogorithm %%%%%%
B = mixgauss_prob(data1, mu1, Sigma1, mixmat1);
path1 = viterbi_path(prior1, transmat1, B);


i really hope someone can help me in this.. i would really appreciate it and once done i will share my code for your reference and future students as well..thanks ...this is my email if wana contact me. premen87@hotmail.com

Thanks

Subject: Speech Recognition via MFCC(feature extraction) and HMM classfication

From: Premen

Date: 20 Sep, 2010 11:28:04

Message: 2 of 8

"Premen " <premen87@hotmail.com> wrote in message <i729rj$7c0$1@fred.mathworks.com>...
> Hai,
>
> currently i doing my thesis(follows as the title) i have done the MFCC as the FE. now i doing the training part and testing phase.
> for training i using the Baum-Welch algorithm for training and viterbi for recognition.
>
> 1)my question is my log-likelihood from my training set is in positive value. i am not sure whether it acceptable o not? most of the value are in positive but in some papers the value was in negative.
>
> 2) i have got some viterbi path. not sure how to make use of it? help me in that too
>
> 3) Actually i am confuse how to do the the recognition part? can anyone willing explain to me. how to do a effective recognition?
>
> My matlab code:
>
> %%%%%%%%Training of HMM (Baum-Welch algorithm)%%%%%%
> [LL, prior1, transmat1, mu1, Sigma1, mixmat1] = ...
> mhmm_em(data1, pi, transmat, mu, Sigma, mixmat, 'max_iter', 5);
> mixmat1 = max(mixmat1, 1.0e-5);
>
> %%%% Calculate the Log Likelihood %%%%%%
> [loglik, errors] = mhmm_logprob(data1, prior1, transmat1, mu1, Sigma1, mixmat1);
> loglik
>
> %%%%%%%%%%%%%% Viterbi Alogorithm %%%%%%
> B = mixgauss_prob(data1, mu1, Sigma1, mixmat1);
> path1 = viterbi_path(prior1, transmat1, B);
>
>
> i really hope someone can help me in this.. i would really appreciate it and once done i will share my code for your reference and future students as well..thanks ...this is my email if wana contact me. premen87@hotmail.com
>
> Thanks



why nobody replying?

Subject: Speech Recognition via MFCC(feature extraction) and HMM classfication

From: Fitri teh

Date: 6 Mar, 2011 15:46:07

Message: 3 of 8

Hye premen,

I'm currently in final years student , my thesis call "Automatic Speech Recognition (ASR) For Speech Therapy" .
My final year project is familiar as you project using features extraction Mel frequency Cepstral Coefficient (MFCC) and Hidden Markov Model (HMM) classification.

Need you help at MFCC and HMM part coding.
I'm using the Baum-Welch algorithm for training and viterbi for recognition.

I didn't understand about this features extraction MFCC and classification HMM.
Need your help to explain it to me ...

Thnks ..
Fitri


"Premen " <premen87@hotmail.com> wrote in message <i729rj$7c0$1@fred.mathworks.com>...
> Hai,
>
> currently i doing my thesis(follows as the title) i have done the MFCC as the FE. now i doing the training part and testing phase.
> for training i using the Baum-Welch algorithm for training and viterbi for recognition.
>
> 1)my question is my log-likelihood from my training set is in positive value. i am not sure whether it acceptable o not? most of the value are in positive but in some papers the value was in negative.
>
> 2) i have got some viterbi path. not sure how to make use of it? help me in that too
>
> 3) Actually i am confuse how to do the the recognition part? can anyone willing explain to me. how to do a effective recognition?
>
> My matlab code:
>
> %%%%%%%%Training of HMM (Baum-Welch algorithm)%%%%%%
> [LL, prior1, transmat1, mu1, Sigma1, mixmat1] = ...
> mhmm_em(data1, pi, transmat, mu, Sigma, mixmat, 'max_iter', 5);
> mixmat1 = max(mixmat1, 1.0e-5);
>
> %%%% Calculate the Log Likelihood %%%%%%
> [loglik, errors] = mhmm_logprob(data1, prior1, transmat1, mu1, Sigma1, mixmat1);
> loglik
>
> %%%%%%%%%%%%%% Viterbi Alogorithm %%%%%%
> B = mixgauss_prob(data1, mu1, Sigma1, mixmat1);
> path1 = viterbi_path(prior1, transmat1, B);
>
>
> i really hope someone can help me in this.. i would really appreciate it and once done i will share my code for your reference and future students as well..thanks ...this is my email if wana contact me. premen87@hotmail.com
>
> Thanks

Subject: Speech Recognition via MFCC(feature extraction) and HMM classfication

From: Premen

Date: 7 Mar, 2011 08:20:09

Message: 4 of 8

"Fitri teh" wrote in message <il0a7v$7i6$1@fred.mathworks.com>...
> Hye premen,
>
> I'm currently in final years student , my thesis call "Automatic Speech Recognition (ASR) For Speech Therapy" .
> My final year project is familiar as you project using features extraction Mel frequency Cepstral Coefficient (MFCC) and Hidden Markov Model (HMM) classification.
>
> Need you help at MFCC and HMM part coding.
> I'm using the Baum-Welch algorithm for training and viterbi for recognition.
>
> I didn't understand about this features extraction MFCC and classification HMM.
> Need your help to explain it to me ...
>
> Thnks ..
> Fitri
>
>

Hi There, Yeah i have done the same topic. MFCC is quite straightforward its just chopping the signal and enhance the frequency. U may google for it. There are a lot.
For HMM, act u can refer some phd or master student project for understanding before you create your own code. Usually University Library should have those references.

Subject: Speech Recognition via MFCC(feature extraction) and HMM classfication

From: deepu.mallapragada@gmail.com

Date: 19 Mar, 2013 11:22:27

Message: 5 of 8

Hi all,
       I need matlab code for MFCC feature extraction. I am doing a project on speaker-diarization. this is my minor-project. As part of my project, I need to use these features, can anyone please mail me the working code for MFCC feature extraction.

Subject: Speech Recognition via MFCC(feature extraction) and HMM classfication

From: Bryant

Date: 19 Apr, 2013 18:50:09

Message: 6 of 8

Have you looked at the VOICEBOX scripts? I believe that it contains codes for MFCC. You can find it by searching MATLAB central or here is the link: http://www.ee.ic.ac.uk/hp/staff/dmb/voicebox/voicebox.html

deepu.mallapragada@gmail.com wrote in message <9943fc63-0257-4815-8841-8e85acc6d45d@googlegroups.com>...
> Hi all,
> I need matlab code for MFCC feature extraction. I am doing a project on speaker-diarization. this is my minor-project. As part of my project, I need to use these features, can anyone please mail me the working code for MFCC feature extraction.

Subject: Speech Recognition via MFCC(feature extraction) and HMM classfication

From: pricky

Date: 20 Mar, 2014 06:17:11

Message: 7 of 8

Hi Bryant
        Please help me out with the process of generating a code book for speech recognition.
Thank You

Subject: Speech Recognition via MFCC(feature extraction) and HMM classfication

From: RITESH

Date: 31 Mar, 2014 10:38:09

Message: 8 of 8

"Premen " <premen87@hotmail.com> wrote in message <i729rj$7c0$1@fred.mathworks.com>...
> Hai,
>
> currently i doing my thesis(follows as the title) i have done the MFCC as the FE. now i doing the training part and testing phase.
> for training i using the Baum-Welch algorithm for training and viterbi for recognition.
>
> 1)my question is my log-likelihood from my training set is in positive value. i am not sure whether it acceptable o not? most of the value are in positive but in some papers the value was in negative.
>
> 2) i have got some viterbi path. not sure how to make use of it? help me in that too
>
> 3) Actually i am confuse how to do the the recognition part? can anyone willing explain to me. how to do a effective recognition?
>
> My matlab code:
>
> %%%%%%%%Training of HMM (Baum-Welch algorithm)%%%%%%
> [LL, prior1, transmat1, mu1, Sigma1, mixmat1] = ...
> mhmm_em(data1, pi, transmat, mu, Sigma, mixmat, 'max_iter', 5);
> mixmat1 = max(mixmat1, 1.0e-5);
>
> %%%% Calculate the Log Likelihood %%%%%%
> [loglik, errors] = mhmm_logprob(data1, prior1, transmat1, mu1, Sigma1, mixmat1);
> loglik
>
> %%%%%%%%%%%%%% Viterbi Alogorithm %%%%%%
> B = mixgauss_prob(data1, mu1, Sigma1, mixmat1);
> path1 = viterbi_path(prior1, transmat1, B);
>
>
> i really hope someone can help me in this.. i would really appreciate it and once done i will share my code for your reference and future students as well..thanks ...this is my email if wana contact me. premen87@hotmail.com
>
> Thanks

If you have the matlab source code for HMM and MFCC can you send me the same.

Tags for this Thread

What are tags?

A tag is like a keyword or category label associated with each thread. Tags make it easier for you to find threads of interest.

Anyone can tag a thread. Tags are public and visible to everyone.

Contact us