Blind stochastic feature transformation for speaker verification over cellular networks

Kwok Kwong Yiu, Man Wai Mak, Ming Cheung Cheung, Sun Yuan Kung

Research output: Chapter in book / Conference proceedingConference article published in proceeding or bookAcademic researchpeer-review

Abstract

Acoustic mismatch between the training and recognition conditions presents one of the serious challenges faced by speaker recognition researchers today. The goal of channel compensation is to achieve performance approaching that of a "matched condition" system while avoiding the need for a large amount of training data. It is important to ensure that the channel compensation algorithms in these systems compensate the channel variation instead of speaker variation. This paper addresses the problem of unsupervised compensation in which the features of a test utterance are transformed to fit the clean speaker model and gender-dependent background model. Specifically, a feature-based transformation is estimated based on the statistical difference between a test utterance and a composite acoustic model formed by combining the speaker and background models. By transforming the features to fit both models, the transformation is implicitly constrained. Experimental results based on the 2001 MIST evaluation set show that the proposed transformation approach achieves significant improvement in both equal error rate and minimum detection cost as compared to cepstral mean subtraction, Znorm and short-time Gaussianization.
Original languageEnglish
Title of host publication2004 International Symposium on Intelligent Multimedia, Video and Speech Processing, ISIMP 2004
Pages679-682
Number of pages4
Publication statusPublished - 1 Dec 2004
Event2004 International Symposium on Intelligent Multimedia, Video and Speech Processing, ISIMP 2004 - Hong Kong, China, Hong Kong
Duration: 20 Oct 200422 Oct 2004

Conference

Conference2004 International Symposium on Intelligent Multimedia, Video and Speech Processing, ISIMP 2004
Country/TerritoryHong Kong
CityHong Kong, China
Period20/10/0422/10/04

Keywords

  • Channel compensation
  • Feature transformation
  • MAP adaptation
  • Robustness
  • Speaker verification

ASJC Scopus subject areas

  • Engineering(all)

Cite this