Název:	Analysis of the influence of speech corpora in the PLDA verification in the task of speaker recognition
Autoři:	Machlica, Lukáš Zají­c, Zbyněk
Citace zdrojového dokumentu:	MACHLICA, Lukáš; ZAJÍC, Zbyněk. Analysis of the influence of speech corpora in the PLDA verification in the task of speaker recognition. In: Text, speech and dialogue. Berlin: Springer, 2012, p. 464-471. (Lectures notes in computer science; 7499). ISBN 978-3-642-32789-6.
Datum vydání:	2012
Nakladatel:	Springer
Typ dokumentu:	článek article
URI:	http://www.kky.zcu.cz/cs/publications/LukasMachlica_2012_Analysisofthe http://hdl.handle.net/11025/17040
ISBN:	978-3-642-32789-6
Klíčová slova:	PLDA;latentní prostor;fůze;supervektor;FA;i-vektor
Klíčová slova v dalším jazyce:	PLDA;latent space;fusion;supervector;FA;i-vector
Abstrakt v dalším jazyce:	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.
Práva:	© Lukáš Machlica - Zbyněk Zajíc
Vyskytuje se v kolekcích:	Články / Articles (KKY) Články / Articles (NTIS)

Všechny záznamy v DSpace jsou chráněny autorskými právy, všechna práva vyhrazena.