A new two-stage scoring normalization approach to speaker verification

Man Wai Mak, W. D. Zhang, M. X. He

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

4 Citations (Scopus)

Abstract

In speaker verification, the cohort and world models have been separately used for scoring normalization. In this work, we embed the two models in elliptical basis function networks and propose a two-stage decision procedure for improving verification performance. The procedure begins with normalization of an utterance by a world model. If the difference between the resulting score and a world threshold is sufficiently large, the claimant is accepted or rejected immediately. Otherwise, the score will be normalized by a cohort model, and the resulting score will be compared with a cohort threshold to make a final accepvreject decision. Experimental evaluations based on the YOHO corpus suggest that the two-stage method achieves a lower ekor rate as compared to the case where only one background model is used.
Original languageEnglish
Title of host publicationProceedings of 2001 International Symposium on Intelligent Multimedia, Video and Speech Processing, ISIMP 2001
Pages107-110
Number of pages4
Publication statusPublished - 1 Dec 2001
Event2001 International Symposium on Intelligent Multimedia, Video and Speech Processing, ISIMP 2001 - Hong Kong, Hong Kong
Duration: 2 May 20014 May 2001

Conference

Conference2001 International Symposium on Intelligent Multimedia, Video and Speech Processing, ISIMP 2001
CountryHong Kong
CityHong Kong
Period2/05/014/05/01

ASJC Scopus subject areas

  • Computer Science(all)

Cite this