Source-specific Informative Prior for i-Vector Extraction

Sven Ewan Shepstone, Kong Aik Lee, Haizhou Li, Zheng-Hua Tan, Søren Holdt Jensen

Research output: Contribution to book/anthology/report/conference proceedingArticle in proceedingResearchpeer-review

5 Citations (Scopus)

Abstract

An i-vector is a low-dimensional fixed-length representation of a variable-length speech utterance, and is defined as the posterior mean of a latent variable conditioned on the observed feature sequence of an utterance. The assumption is that the prior for the latent variable is non-informative, since for homogeneous datasets there is no gain in generality in using an informative prior. This work shows that extracting i-vectors for a heterogeneous dataset, containing speech samples recorded from multiple sources, using informative priors instead is applicable, and leads to favorable results. Tests carried out on the NIST 2008 and 2010 Speaker Recognition Evaluation (SRE) dataset show that our proposed method beats three baselines: For the short2-short3 core-task in SRE'08, for the female and male cases, five and six respectively, out of eight common conditions were beaten, and for the core-core task in SRE'10, for both genders, five out of nine common conditions were beaten.
Original languageEnglish
Title of host publicationIEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), 2015
PublisherIEEE Signal Processing Society
Publication dateApr 2015
Pages4185 - 4189
ISBN (Electronic)978-1-4673-6997-8
DOIs
Publication statusPublished - Apr 2015
Event40th IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP) 2015 - Brisbane, Australia
Duration: 19 Apr 201524 Apr 2015
Conference number: 2015

Conference

Conference40th IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP) 2015
Number2015
CountryAustralia
CityBrisbane
Period19/04/201524/04/2015
SeriesI E E E International Conference on Acoustics, Speech and Signal Processing. Proceedings
ISSN1520-6149

Fingerprint Dive into the research topics of 'Source-specific Informative Prior for i-Vector Extraction'. Together they form a unique fingerprint.

Cite this