Conference paper

KARAFIÁT Martin, BASKAR Murali K., MATĚJKA Pavel, VESELÝ Karel, GRÉZL František and ČERNOCKÝ Jan. Multilingual BLSTM and Speaker-Specific Vector Adaptation in 2016 BUT BABEL SYSTEM. In: Proceedings of SLT 2016. San Diego: IEEE Signal Processing Society, 2016, pp. 637-643. ISBN 978-1-5090-4903-5.
Publication language:english
Original title:Multilingual BLSTM and Speaker-Specific Vector Adaptation in 2016 BUT BABEL SYSTEM
Title (cs):Multilingvální BLSTM a adaptace pomocí vektorů specifických pro řečníka ve VUT Babel 2016 systému
Pages:637-643
Proceedings:Proceedings of SLT 2016
Conference:2016 IEEE Workshop on Spoken Language Technology
Place:San Diego, US
Year:2016
ISBN:978-1-5090-4903-5
Publisher:IEEE Signal Processing Society
URL:http://www.fit.vutbr.cz/research/groups/speech/publi/2016/karafiat_slt2016_0000637-1.pdf [PDF]
Files: 
+Type Name Title Size Last modified
iconkarafiat_slt2016_0000637-1.pdf218 KB2017-03-06 11:03:53
^ Select all
With selected:
Keywords
Automatic speech recognition, Multilingual neural networks, Bidirectional Long Short Term Memory, i-vectors, Sequence Summarizing Neural Networks.
Annotation
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.
Abstract
This paper provides an extensive summary of BUT 2016 system for the last IARPA Babel evaluations. It concentrates on multi-lingual training of both deep neural network (DNN)-based feature extraction and acoustic models including multilingual training of bidirectional Long Short Term memory networks. Next, two low-dimensional vector approaches to speaker adaptation are investigated: i-vectors and sequence-summarizing neural networks (SSNN). The results provided on three Babel Year 4 languages show clear advantage of both approaches in case limited amount of training data is available. The time necessary for the development of a new system is addressed too, as some of the investigated techniques do not require extensive re-training of the whole system.
BibTeX:
@INPROCEEDINGS{
   author = {Martin Karafi{\'{a}}t and K. Murali Baskar and Pavel
	Mat{\v{e}}jka and Karel Vesel{\'{y}} and Franti{\v{s}}ek
	Gr{\'{e}}zl and Jan {\v{C}}ernock{\'{y}}},
   title = {Multilingual BLSTM and Speaker-Specific Vector Adaptation in
	2016 BUT BABEL SYSTEM},
   pages = {637--643},
   booktitle = {Proceedings of SLT 2016},
   year = {2016},
   location = {San Diego, US},
   publisher = {IEEE Signal Processing Society},
   ISBN = {978-1-5090-4903-5},
   language = {english},
   url = {http://www.fit.vutbr.cz/research/view_pub.php.en.iso-8859-2?id=11310}
}

Your IPv4 address: 54.158.55.5
Switch to IPv6 connection

DNSSEC [dnssec]