STBU System for the NIST 2006 Speaker Recognition Evaluation

Author

Matejka, R. ; Burget, Lukas ; Schwarz, P. ; Glembek, O. ; Karafiat, Martin ; Grezl, Frantisek ; Cernocky, Jan ; van Leeuwen, David A. ; Brummer, N. ; Strasheim, A.

Author_Institution

Fac. of Inf. Technol., Brno Univ. of Technol., Czech Republic

Volume

4

fYear

2007

fDate

15-20 April 2007

Abstract

This paper describes STBU 2006 speaker recognition system, which performed well in the NIST 2006 speaker recognition evaluation. STBU is consortium of 4 partners: Spescom DataVoice (South Africa), TNO (Netherlands), BUT (Czech Republic) and University of Stellenbosch (South Africa). The primary system is a combination of three main kinds of systems: (1) GMM, with short-time MFCC or PLP features, (2) GMM-SVM, using GMM mean supervectors as input and (3) MLLR-SVM, using MLLR speaker adaptation coefficients derived from English LVCSR system. In this paper, we describe these sub-systems and present results for each system alone and in combination on the NIST Speaker Recognition Evaluation (SRE) 2006 development and evaluation data sets.

Keywords

Gaussian processes; speaker recognition; support vector machines; English LVCSR system; GMM; GMM mean supervectors; GMM-SVM; MLLR-SVM; NIST 2006 Speaker Recognition Evaluation; PLP features; STBU system; short-time MFCC; speaker adaptation coefficients; Africa; Gaussian processes; Maximum likelihood linear regression; Mel frequency cepstral coefficient; NIST; Speaker recognition; Speech recognition; Sun; Support vector machines; Testing; GMM; NAP; SVM; Speaker recognition; eigenchannel;

fLanguage

English

Publisher

ieee

Conference_Titel

Acoustics, Speech and Signal Processing, 2007. ICASSP 2007. IEEE International Conference on

Conference_Location

Honolulu, HI

ISSN

1520-6149

Print_ISBN

1-4244-0727-3

Type

conf

DOI

10.1109/ICASSP.2007.367203

Filename

4218077