Using Fishervoice to enhance the performance of I-vector based speaker verification system

Na Li, Xiangyang Zeng, Zhifeng Li, Yu Qiao, Weiwu Jiang

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

Abstract

I-vector is a popular feature representation technique in speaker verification systems. In this paper, we use Fishervoice algorithm in combination with i-vector feature representation to improve speaker verification performance. By applying the Fishervoice model to map the i-vector into a low-dimensional discriminant subspace, the intra-speaker variability can be reduced and the discriminative class boundary information can be emphasized for enhanced recognition performance. Experiments on NIST SRE 2008 core test task show that the proposed framework achieves 19.9% and 8.5% dramatic relative decrease in EER and minDCF metrics respectively compared to the state-of-the-art PLDA based method.

Original languageEnglish
Title of host publicationICIST 2014 - Proceedings of 2014 4th IEEE International Conference on Information Science and Technology
EditorsGuoqing Xu, Yu Qiao, Xinyu Wu
PublisherInstitute of Electrical and Electronics Engineers Inc.
Pages578-581
Number of pages4
ISBN (Electronic)9781479948086
DOIs
StatePublished - 10 Oct 2014
Event2014 4th IEEE International Conference on Information Science and Technology, ICIST 2014 - Shenzhen, China
Duration: 26 Apr 201428 Apr 2014

Publication series

NameICIST 2014 - Proceedings of 2014 4th IEEE International Conference on Information Science and Technology

Conference

Conference2014 4th IEEE International Conference on Information Science and Technology, ICIST 2014
Country/TerritoryChina
CityShenzhen
Period26/04/1428/04/14

Keywords

  • discriminative
  • Fishervoice
  • i-vector
  • PLDA

Fingerprint

Dive into the research topics of 'Using Fishervoice to enhance the performance of I-vector based speaker verification system'. Together they form a unique fingerprint.

Cite this