Detail publikace

Multilingual BLSTM and Speaker-Specific Vector Adaptation in 2016 BUT BABEL SYSTEM

KARAFIÁT, M. BASKAR, M. MATĚJKA, P. VESELÝ, K. GRÉZL, F. ČERNOCKÝ, J.

Originální název

Multilingual BLSTM and Speaker-Specific Vector Adaptation in 2016 BUT BABEL SYSTEM

Typ

článek ve sborníku ve WoS nebo Scopus

Jazyk

angličtina

Originální abstrakt

This paper provides an extensive summary of BUT 2016 system for the last Babel evaluations. It concentrates on multi-lingual training of both DNN-based features and acoustic models and on the lowdimensional to speaker adaptation.

Klíčová slova

Automatic speech recognition, Multilingual neural networks, Bidirectional Long Short Term Memory, i-vectors, Sequence Summarizing Neural Networks.

Autoři

KARAFIÁT, M.; BASKAR, M.; MATĚJKA, P.; VESELÝ, K.; GRÉZL, F.; ČERNOCKÝ, J.

Vydáno

13. 12. 2016

Nakladatel

IEEE Signal Processing Society

Místo

San Diego

ISBN

978-1-5090-4903-5

Kniha

Proceedings of SLT 2016

Strany od

637

Strany do

643

Strany počet

7

URL

https://www.fit.vut.cz/research/publication/11310/

BibTex

@inproceedings{BUT132604,
  author="Martin {Karafiát} and Murali Karthick {Baskar} and Pavel {Matějka} and Karel {Veselý} and František {Grézl} and Jan {Černocký}",
  title="Multilingual BLSTM and Speaker-Specific Vector Adaptation in 2016 BUT BABEL SYSTEM",
  booktitle="Proceedings of SLT 2016",
  year="2016",
  pages="637--643",
  publisher="IEEE Signal Processing Society",
  address="San Diego",
  doi="10.1109/SLT.2016.7846330",
  isbn="978-1-5090-4903-5",
  url="https://www.fit.vut.cz/research/publication/11310/"
}