Thumbnail
Access Restriction
Subscribed

Author Kenny, P. ♦ Boulianne, G. ♦ Ouellet, P. ♦ Dumouchel, P.
Sponsorship IEEE Signal Processing Society
Source IEEE Xplore Digital Library
Content type Text
Publisher Institute of Electrical and Electronics Engineers, Inc. (IEEE)
File Format PDF
Copyright Year ©2006
Language English
Subject Domain (in DDC) Natural sciences & mathematics ♦ Physics ♦ Electricity & electronics
Subject Keyword Speaker recognition ♦ Testing ♦ NIST ♦ Performance analysis ♦ Adaptation model ♦ Speech recognition ♦ Large-scale systems ♦ Statistical analysis ♦ Statistical distributions ♦ Performance evaluation ♦ speaker verification ♦ Factor analysis ♦ Gaussian mixture
Abstract We present a corpus-based approach to speaker verification in which maximum-likelihood II criteria are used to train a large-scale generative model of speaker and session variability which we call joint factor analysis. Enrolling a target speaker consists in calculating the posterior distribution of the hidden variables in the factor analysis model and verification tests are conducted using a new type of likelihood II ratio statistic. Using the NIST 1999 and 2000 speaker recognition evaluation data sets, we show that the effectiveness of this approach depends on the availability of a training corpus which is well matched with the evaluation set used for testing. Experiments on the NIST 1999 evaluation set using a mismatched corpus to train factor analysis models did not result in any improvement over standard methods, but we found that, even with this type of mismatch, feature warping performs extremely well in conjunction with the factor analysis model, and this enabled us to obtain very good results (equal error rates of about 6.2%)
Description Author affiliation :: Centre de Recherche Informatique de Montreal, Que.
ISSN 15587916
Education Level UG and PG
Learning Resource Type Article
Publisher Date 2007-05-01
Publisher Place U.S.A.
Rights Holder Institute of Electrical and Electronics Engineers, Inc. (IEEE)
Volume Number 15
Issue Number 4
Size (in Bytes) 502.64 kB
Page Count 13
Starting Page 1448
Ending Page 1460


Source: IEEE Xplore Digital Library