Publication Details
Multilingual Bottleneck Features for Language Recognition
Matějka Pavel, Ing., Ph.D. (DCGM FIT BUT)
Grézl František, Ing., Ph.D. (DCGM FIT BUT)
Plchot Oldřich, Ing., Ph.D. (DCGM FIT BUT)
Černocký Jan, prof. Dr. Ing. (DCGM FIT BUT)
multilingual training, stacked bottleneck features, language identification
In this work, we applied multilingual training paradigm of SBN neural networks to extract linguistically rich features.
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
@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" }