A hybrid GMM/SVM approach to speaker identification

Author

Fine, Shai ; Navratil, J. ; Gopinath, Ramesh A.

Author_Institution

IBM Thomas J. Watson Res. Center, Yorktown Heights, NY, USA

Volume

1

fYear

2001

fDate

2001

Firstpage

417

Abstract

Proposes a classification scheme that incorporates statistical models and support vector machines. A hybrid system which appropriately combines the advantages of both the generative and discriminant model paradigms is described and experimentally evaluated on a text-independent speaker recognition task in matched and mismatched training and test conditions. Our results prove that the combination is beneficial in terms of performance and practical in terms of computation. We report relative improvements of up to 25% reduction in identification error rate compared to the baseline statistical model

Keywords

covariance matrices; information theory; learning (artificial intelligence); learning automata; pattern classification; speaker recognition; statistical analysis; classification scheme; discriminant model; generative model; hybrid GMM/SVM approach; hybrid Gaussian mixture models/support vector machine approach; identification error rate; speaker identification; statistical models; text-independent speaker recognition task; Acoustic testing; Electronic mail; Hidden Markov models; Hybrid power systems; Kernel; Power generation; Robustness; Speaker recognition; Support vector machine classification; Support vector machines;

fLanguage

English

Publisher

ieee

Conference_Titel

Acoustics, Speech, and Signal Processing, 2001. Proceedings. (ICASSP '01). 2001 IEEE International Conference on

Conference_Location

Salt Lake City, UT

ISSN

1520-6149

Print_ISBN

0-7803-7041-4

Type

conf

DOI

10.1109/ICASSP.2001.940856

Filename

940856