A new method for fuzzy hidden Markov models in speech recognition

Author

Tarihi, Mohammad Reza ; Taheri, Asghar ; Bababeyk, Hassan

fYear

2005

fDate

17-18 Sept. 2005

Firstpage

36

Lastpage

40

Abstract

This paper proposes a fuzzy approach to the hidden Markov model (HMM). This method celled the fuzzy HMM for speech and speaker recognition as an application of fuzzy expectation maximizing algorithm in HMM. The fuzzy HMM algorithm is regarded as an application of the fuzzy expectation-maximization (EM) algorithm to the Baum-Welch algorithm in the HMM. The Texas Instruments p4 used speech and speaker recognition experiments and show better results for fuzzy HMMs compared with conventional HMMs. Equation and how estimation of discrete and continuous HMM parameters on based this two algorithm is explained and performance of two speech recognition method for one hundred is surveyed. This paper show better results for the fuzzy HMM, compared with the conventional HMM.

Keywords

expectation-maximisation algorithm; fuzzy set theory; hidden Markov models; speaker recognition; Baum-Welch algorithm; fuzzy expectation-maximization algorithm; fuzzy hidden Markov models; speaker recognition; speech recognition method; Clustering algorithms; Electronic mail; Hidden Markov models; Instruments; Parameter estimation; Probability distribution; Speaker recognition; Speech recognition; Stochastic processes; Vector quantization;

fLanguage

English

Publisher

ieee

Conference_Titel

Emerging Technologies, 2005. Proceedings of the IEEE Symposium on

Print_ISBN

0-7803-9247-7

Type

conf

DOI

10.1109/ICET.2005.1558851

Filename

1558851