Title | ||
---|---|---|
GMM-HMM acoustic model training by a two level procedure with Gaussian components determined by automatic model selection |
Abstract | ||
---|---|---|
This paper investigates the Bayesian Ying-Yang (BYY) learning for speech recognition via Gaussian mixture models (GMMs) based Hidden Markov models (HMMs). A two level procedure is proposed with the hidden Markov level trained still under the maximum likelihood principle by the Baum-Welch algorithm but with the GMMs level trained under the BYY best harmony. We proposed a new batch way EM-like Ying-Yang alternation algorithm and used it as a plug-in block to the Baum-Welch algorithm. The advantage is that number of GMM components can be automatically determined during this BYY harmony learning and that the resulted model parameters become less affected than EM-ML training by the problem of overfitting and singular solution. In comparison with the standard EM-ML training and classical model selection criterions, including BIC and AIC, speech recognition experiments in a large vocabulary task on the Hub4 broadcast news database shown that the proposed algorithm provides an improved performance and also good convergence. |
Year | DOI | Venue |
---|---|---|
2010 | 10.1109/ICASSP.2010.5495122 | ICASSP |
Keywords | Field | DocType |
vocabulary task,speech recognition,hidden markov model,gmm-hmm acoustic model training,bayesian ying-yang learning,hmms,hub4 broadcast,model selection,bayes methods,learning (artificial intelligence),two level procedure,gmms,baum-welch algorithm,acoustic signal processing,automatic model selection,hidden markov models,gaussian mixture model,speech,baum welch algorithm,maximum likelihood estimation,bayesian methods,convergence,automatic speech recognition,baum welch,acoustics,learning artificial intelligence,databases,singular solution | Pattern recognition,Computer science,Model selection,Speech recognition,Gaussian,Artificial intelligence,Overfitting,Hidden Markov model,Baum–Welch algorithm,Mixture model,Acoustic model,Bayesian probability | Conference |
ISSN | ISBN | Citations |
1520-6149 E-ISBN : 978-1-4244-4296-6 | 978-1-4244-4296-6 | 8 |
PageRank | References | Authors |
0.57 | 4 | 3 |