A SVM/HMM system for speaker recognition

Author

Campbell, W.M.

Author_Institution

Motorola Human Interface Lab, Tempe, AZ, USA

Volume

fYear

2003

fDate

6-10 April 2003

Abstract

A framework for combining support vector machines with hidden Markov models (HMM) is given. A HMM is used with a Viterbi alignment to generate a set of subsequences of feature vectors. Each subsequence is then scored using a support vector machine sequence kernel. Experiments are performed for both text-independent and text-prompted speaker recognition tasks. Results show that the method can dramatically reduce error rates over a support vector machine (SVM) only system.

Keywords

error statistics; feature extraction; hidden Markov models; learning automata; maximum likelihood estimation; sequences; speaker recognition; SVM/HMM system; Viterbi alignment; error rates; feature vector subsequence; hidden Markov models; sequence kernel; speaker recognition; support vector machines; text-independent speaker recognition; text-prompted speaker recognition; Error analysis; Hidden Markov models; Humans; Kernel; Scalability; Speaker recognition; Speech processing; Support vector machine classification; Support vector machines; Viterbi algorithm;

fLanguage

English

Publisher

ieee

Conference_Titel

Acoustics, Speech, and Signal Processing, 2003. Proceedings. (ICASSP '03). 2003 IEEE International Conference on

ISSN

1520-6149

Print_ISBN

0-7803-7663-3

Type

conf

DOI

10.1109/ICASSP.2003.1202331

Filename

1202331

Link To Document

https://search.isc.ac/dl/search/defaultta.aspx?DTC=49&DC=3489098