Conference paper

BRÜMMER Niko, STRASHEIM Albert, HUBEIKA Valiantsina, MATĚJKA Pavel, BURGET Lukáš and GLEMBEK Ondřej. Discriminative Acoustic Language Recognition via Channel-Compensated GMM Statistics. In: Proc. Interspeech 2009. Brighton: International Speech Communication Association, 2009, pp. 2187-2190. ISBN 978-1-61567-692-7. ISSN 1990-9772.
Publication language:english
Original title:Discriminative Acoustic Language Recognition via Channel-Compensated GMM Statistics
Title (cs):Diskriminativní akustické rozpoznávání jazyka pomocí kanálově kompenzovaných GMM statistik
Pages:2187-2190
Proceedings:Proc. Interspeech 2009
Conference:Interspeech 2009
Place:Brighton, GB
Year:2009
ISBN:978-1-61567-692-7
Journal:Proceedings of Interspeech, No. 9, FR
ISSN:1990-9772
Publisher:International Speech Communication Association
URL:http://www.fit.vutbr.cz/research/groups/speech/publi/2009/brummer_is2009.pdf [PDF]
Keywords
acoustic language recognition, intersession variability compensation, discriminative training
Annotation
The paper is on Discriminative Acoustic Language Recognition via Channel-Compensated GMM Statistics. The results are reported on NIST LRE'07.
Abstract
We propose a novel design for acoustic feature-based automatic spoken language recognizers. Our design is inspired by recent advances in text-independent speaker recognition, where intraclass variability is modeled by factor analysis in Gaussian mixture model (GMM) space. We use approximations to GMMlikelihoods which allow variable-length data sequences to be represented as statistics of fixed size. Our experiments on NIST LRE'07 show that variability-compensation of these statistics can reduce error-rates by a factor of three. Finally, we show that further improvements are possible with discriminative logistic regression training.
BibTeX:
@INPROCEEDINGS{
   author = {Niko Br{\"{u}}mmer and Albert Strasheim and Valiantsina
	Hubeika and Pavel Mat{\v{e}}jka and Luk{\'{a}}{\v{s}} Burget
	and Ond{\v{r}}ej Glembek},
   title = {Discriminative Acoustic Language Recognition via
	Channel-Compensated GMM Statistics},
   pages = {2187--2190},
   booktitle = {Proc. Interspeech 2009},
   journal = {Proceedings of Interspeech},
   number = {9},
   year = {2009},
   location = {Brighton, GB},
   publisher = {International Speech Communication Association},
   ISBN = {978-1-61567-692-7},
   ISSN = {1990-9772},
   language = {english},
   url = {http://www.fit.vutbr.cz/research/view_pub.php?id=9042}
}

Your IPv4 address: 54.156.92.243
Switch to IPv6 connection

DNSSEC [dnssec]