Skip to content

Publications

Detail of publication

Citation

Lukáš Machlica and Zbyněk Zajíc : Analysis of the Influence of Speech Corpora in the PLDA Verification in the Task of Speaker Recognition . Lecture Notes in Computer Science, p. 464-471, 2012.

Download PDF

PDF

Abstract

In the paper recent methods used in the task of speaker recognition are presented. At first, the extraction of so called i-vectors from GMM based supervectors is discussed. These i-vectors are of low dimension and lie in a subspace denoted as Total Variability Space (TVS). The focus of the paper is put on Probabilistic Linear Discriminant Analysis (PLDA), which is used as a generative model in the TVS. The influence of development data is analyzed utilizing distinct speech corpora. It is shown that it is preferable to cluster available speech corpora to classes, train one PLDA model for each class and fuse the results at the end. Experiments are presented on NIST Speaker Recognition Evaluation (SRE) 2008 and NIST SRE 2010.

Detail of publication

Title: Analysis of the Influence of Speech Corpora in the PLDA Verification in the Task of Speaker Recognition
Author: Lukáš Machlica ; Zbyněk Zajíc
Language: English
Date of publication: 3 Jul 2012
Year: 2012
Type of publication: Papers in journals
Title of journal or book: Lecture Notes in Computer Science
Page: 464 - 471
DOI: 10.1007/978-3-642-32790-2_56
ISSN: 0302-9743
/ 2013-09-06 14:05:45 /

Keywords

PLDA, latent space, fusion, supervector, FA, i-vector

BibTeX

@MISC{LukasMachlica_2012_Analysisofthe,
 author = {Luk\'{a}\v{s} Machlica and Zbyn\v{e}k Zaj\'{i}c},
 title = {Analysis of the Influence of Speech Corpora in the PLDA Verification in the Task of Speaker Recognition},
 year = {2012},
 journal = {Lecture Notes in Computer Science},
 pages = {464-471},
 ISSN = {0302-9743},
 doi = {10.1007/978-3-642-32790-2_56 },
 url = {http://www.kky.zcu.cz/en/publications/LukasMachlica_2012_Analysisofthe},
}