Publication Details

Multilingual Bottleneck Features for Language Recognition

FÉR Radek, MATĚJKA Pavel, GRÉZL František, PLCHOT Oldřich and ČERNOCKÝ Jan. Multilingual Bottleneck Features for Language Recognition. In: Proceedings of Interspeech 2015. Dresden: International Speech Communication Association, 2015, pp. 389-393. ISBN 978-1-5108-1790-6. ISSN 1990-9772.
Czech title
Multilingvální příznaky z neuronové sítě s úzkým hrdlem pro rozpoznávání jazyka
Type
conference paper
Language
english
Authors
URL
Keywords

multilingual training, stacked bottleneck features, language identification

Abstract

In this work, we applied multilingual training paradigm of SBN neural networks to extract linguistically rich features.

Annotation

In this paper, we investigate Multilingual Stacked Bottleneck Features (SBN) in language recognition domain. These features are extracted using bottleneck neural networks trained on data from multiple languages. Previous results have shown benefits of multilingual training of SBN feature extractor for speech recognition. Here we focus on its impact on language recognition. We present results obtained with monolingual and multilingual networks, and their fusions. Using multilingual features, we obtain 16% relative improvement on 3 s condition of NIST LRE09 dataset with respect to features trained on a single language

Published
2015
Pages
389-393
Journal
Proceedings of Interspeech - on-line, vol. 2015, no. 9, ISSN 1990-9772
Proceedings
Proceedings of Interspeech 2015
Conference
Interspeech Conference, Dresden, DE
ISBN
978-1-5108-1790-6
Publisher
International Speech Communication Association
Place
Dresden, DE
UT WoS
000380581600079
EID Scopus
BibTeX
@INPROCEEDINGS{FITPUB10966,
   author = "Radek F\'{e}r and Pavel Mat\v{e}jka and Franti\v{s}ek Gr\'{e}zl and Old\v{r}ich Plchot and Jan \v{C}ernock\'{y}",
   title = "Multilingual Bottleneck Features for Language Recognition",
   pages = "389--393",
   booktitle = "Proceedings of Interspeech 2015",
   journal = "Proceedings of Interspeech - on-line",
   volume = 2015,
   number = 09,
   year = 2015,
   location = "Dresden, DE",
   publisher = "International Speech Communication Association",
   ISBN = "978-1-5108-1790-6",
   ISSN = "1990-9772",
   language = "english",
   url = "https://www.fit.vut.cz/research/publication/10966"
}
Back to top