An approach to robust speaker recognition using stochastic matching

Author

Ma, Jiyong ; Gao, Wen

Author_Institution

Inst. of Comput. Technol., Acad. Sinica, Beijing, China

Volume

2

fYear

2000

fDate

2000

Firstpage

803

Abstract

This paper describes an approach to robust speaker recognition using stochastic matching for depressing channel mismatch in telephone networks. For the two recognition algorithms such as GMM (Gaussian mixture models) for text-independent speaker identification and HMM for text prompted speaker verification, the corresponding stochastic matching approaches are presented in detail. Experimental results have shown that the stochastic matching approach for channel compensation based on the compensation term from GMM is better than that of the compensation term from a single Gaussian model

Keywords

Gaussian processes; compensation; hidden Markov models; speaker recognition; telecommunication channels; telephone networks; GMM; Gaussian mixture models; HMM; channel compensation; channel mismatch; compensation term; robust speaker recognition; stochastic matching; telephone networks; text prompted speaker verification; text-independent speaker identification; Cepstral analysis; Equations; Hidden Markov models; Robustness; Speaker recognition; Speech enhancement; Stochastic processes; Stochastic resonance; Telephony; Working environment noise;

fLanguage

English

Publisher

ieee

Conference_Titel

Signal Processing Proceedings, 2000. WCCC-ICSP 2000. 5th International Conference on

Conference_Location

Beijing

Print_ISBN

0-7803-5747-7

Type

conf

DOI

10.1109/ICOSP.2000.891632

Filename

891632