Publication Details

Discriminative Training and Channel Compensation for Acoustic Language Recognition

HUBEIKA Valiantsina, BURGET Lukáš, MATĚJKA Pavel and SCHWARZ Petr. Discriminative Training and Channel Compensation for Acoustic Language Recognition. In: Proc. Interspeech 2008. Brisbane: International Speech Communication Association, 2008, p. 4. ISSN 1990-9772.
Czech title
Diskriminativní trénování a kompenzace kanálu pro akustické rozpoznávání jazyka
Type
conference paper
Language
english
Authors
Hubeika Valiantsina, Ing. (DCGM FIT BUT)
Burget Lukáš, doc. Ing., Ph.D. (DCGM FIT BUT)
Matějka Pavel, Ing., Ph.D. (DCGM FIT BUT)
Schwarz Petr, Ing., Ph.D. (DCGM FIT BUT)
URL
Keywords

language recognition

Abstract

The paper is on Discriminative Training and Channel Compensation for Acoustic Language Recognition

Annotation

This paper describes the acoustic language recognition subsystems of Brno University of Technology (BUT) which contributed to the BUT main submission to the NIST LRE 2007. Two main techniques are employed in the subsystems discriminative training in terms of Maximum Mutual Information, and channel compensation in terms of eigenchannel adaptation in both, model and feature domain. The complementarity of the approaches is analyzed.

Published
2008
Pages
4
Journal
Proceedings of Interspeech - on-line, no. 9, ISSN 1990-9772
Proceedings
Proc. Interspeech 2008
Conference
Interspeech Conference, Brisbane, AU
Publisher
International Speech Communication Association
Place
Brisbane, AU
BibTeX 
@INPROCEEDINGS{FITPUB8726,
   author = "Valiantsina Hubeika and Luk\'{a}\v{s} Burget and Pavel Mat\v{e}jka and Petr Schwarz",
   title = "Discriminative Training and Channel Compensation for Acoustic Language Recognition",
   pages = 4,
   booktitle = "Proc. Interspeech 2008",
   journal = "Proceedings of Interspeech - on-line",
   number = 9,
   year = 2008,
   location = "Brisbane, AU",
   publisher = "International Speech Communication Association",
   ISSN = "1990-9772",
   language = "english",
   url = "https://www.fit.vut.cz/research/publication/8726"
}
Back to top