A fuzzy approach to statistical models in speech and speaker recognition

Author

Tran, Dat ; Wagner, Michael ; Zheng, Tongtao

Author_Institution

Sch. of Comput., Canberra Univ., ACT., Australia

Volume

3

fYear

1999

fDate

22-25 Aug. 1999

Firstpage

1275

Abstract

A unified fuzzy approach to statistical models for speech and speaker recognition is presented. Since the expectation-maximisation (EM) algorithm is a powerful learning method for maximising the likelihood of the observed data in the presence of hidden variables, the fuzzy EM algorithm based on the fuzzy c-means algorithm is thereby established. From this fuzzy EM algorithm, the fuzzy algorithms for hidden Markov models, Gaussian mixture models, and vector quantisation are developed. The experimental results on T146 and ANDOSL speech data corpora for speech and speaker recognition show that the fuzzy approach is capable of achieving higher recognition accuracy.

Keywords

fuzzy set theory; hidden Markov models; maximum likelihood estimation; speech recognition; statistical analysis; vector quantisation; Gaussian mixture models; expectation-maximisation algorithm; fuzzy EM algorithm; fuzzy c-means algorithm; fuzzy set theory; hidden Markov models; learning method; maximum likelihood estimation; speaker recognition; speech recognition; statistical models; vector quantisation; Algorithm design and analysis; Clustering algorithms; Hidden Markov models; Iterative algorithms; Maximum likelihood estimation; Parameter estimation; Partitioning algorithms; Speaker recognition; Speech; Vector quantization;

fLanguage

English

Publisher

ieee

Conference_Titel

Fuzzy Systems Conference Proceedings, 1999. FUZZ-IEEE '99. 1999 IEEE International

Conference_Location

Seoul, South Korea

ISSN

1098-7584

Print_ISBN

0-7803-5406-0

Type

conf

DOI

10.1109/FUZZY.1999.790085

Filename

790085