Full-covariance UBM and heavy-tailed PLDA in i-vector speaker verification

In this paper, we describe recent progress in i-vector based speaker verification. The use of universal background models (UBM) with full-covariance matrices is suggested and thoroughly experimentally tested. The i-vectors are scored using a simple cosine distance and advanced techniques such as Pro...

Full description

Saved in:
Bibliographic Details
Main Authors: Matejka, Pavel, Glembek, Ondfej, Castaldo, Fabio, Alam, M. J., Plchot, Oldfich, Kenny, Patrick, Burget, Lukas, Cernocky, Jan
Format: Conference Proceeding
Language:English
Subjects:
Covariance matrix
Feature extraction
full-covariance UBM
GMM
heavy-tailed PLDA
i-vectors
NIST
PLDA
Speaker recognition
Speech
Speech recognition
Training
Online Access:Request full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:In this paper, we describe recent progress in i-vector based speaker verification. The use of universal background models (UBM) with full-covariance matrices is suggested and thoroughly experimentally tested. The i-vectors are scored using a simple cosine distance and advanced techniques such as Probabilistic Linear Discriminant Analysis (PLDA) and heavy-tailed variant of PLDA (PLDA-HT). Finally, we investigate into dimensionality reduction of i-vectors be fore entering the PLDA-HT modeling. The results are very competitive: on NIST 2010 SRE task, the results of a single full-covariance LDA-PLDA-HT system approach those of complex fused system.
ISSN:1520-6149
2379-190X
DOI:10.1109/ICASSP.2011.5947436