GMM and kernel-based speaker recognition with the ISIP toolkit

Author

Imbiriba, T. ; Klautau, Aldeharo ; Pariha, Naveen ; Raghavan, Sridhar ; Picone, Joseph

Author_Institution

Signal Process. Lab., Universidade Federal do Para

fYear

2004

fDate

Sept. 29 2004-Oct. 1 2004

Firstpage

371

Lastpage

380

Abstract

This paper describes an open source framework for developing speaker recognition systems. Among other features, it supports kernel classifiers, such as the support and relevance vector machines. The paper also presents results for the IME corpus using Gaussian mixture models, which outperforms previously published ones, and discusses strategies for applying discriminative classifiers to speaker recognition

Keywords

Gaussian processes; speaker recognition; support vector machines; GMM-based speaker recognition; Gaussian mixture models; IME corpus; ISIP toolkit; discriminative classifiers; kernel classifiers; kernel-based speaker recognition; open source framework; relevance vector machines; support vector machines; Computer architecture; Hidden Markov models; Kernel; Maximum likelihood linear regression; Production systems; Signal processing; Speaker recognition; Speech recognition; Support vector machine classification; Support vector machines;

fLanguage

English

Publisher

ieee

Conference_Titel

Machine Learning for Signal Processing, 2004. Proceedings of the 2004 14th IEEE Signal Processing Society Workshop

Conference_Location

Sao Luis

ISSN

1551-2541

Print_ISBN

0-7803-8608-4

Type

conf

DOI

10.1109/MLSP.2004.1422996

Filename

1422996