Conference paper

KARAFIÁT Martin, SZŐKE Igor and ČERNOCKÝ Jan. Using Gradient Descent Optimization for Acoustics Training from Heterogeneous Data. In: Proc. Text, Speech and Dialog 2010. Brno: Springer Verlag, 2010, pp. 322-329. ISBN 978-3-642-15759-2. ISSN 0302-9743.
Publication language:english
Original title:Using Gradient Descent Optimization for Acoustic Training from Heterogeneous Data
Title (cs):Využití gradient descent optimalizace pro trénování akustických modelů z heterogenních dat
Pages:322-329
Proceedings:Proc. Text, Speech and Dialog 2010
Conference:13th International Conference on Text, Speech and Dialogue, TSD 2010
Series:LNAI 6231
Place:Brno, CZ
Year:2010
ISBN:978-3-642-15759-2
Journal:Lecture Notes in Computer Science, Vol. 2010, No. 9, DE
ISSN:0302-9743
Publisher:Springer Verlag
URL:http://www.fit.vutbr.cz/research/groups/speech/publi/2010/karafiat_TSD_2010_322.pdf [PDF]
Keywords
speech, acoustic models, heterogeneous data, HLDA system, gradient descent training, robustness
Annotation
This paper is on using the gradient descent optimization for acoustics training from heterogeneous data. We study the use of heterogeneous data for training of acoustic models.
Abstract
In this paper, we study the use of heterogeneous data for training of acoustic models. In initial experiments, a significant drop of accuracy has been observed on in-domain test set if the data was added without any regularization. A solution is proposed by getting control over the training data by optimization of the weights of different data-sets. The final models shows good performance on all various tests linked to various speaking styles. Furthermore, we used this approach to increase the performance over just the main test set. We obtained 0.3% absolute improvement on basic system and 0.4% on HLDA system although the size of the heterogeneous data set was quite small.
BibTeX:
@INPROCEEDINGS{
   author = {Martin Karafiát and Igor Szőke and Jan Černocký},
   title = {Using Gradient Descent Optimization for Acoustic Training
	from Heterogeneous Data},
   pages = {322--329},
   booktitle = {Proc. Text, Speech and Dialog 2010},
   series = {LNAI 6231},
   journal = {Lecture Notes in Computer Science},
   volume = {2010},
   number = {9},
   year = {2010},
   location = {Brno, CZ},
   publisher = {Springer Verlag},
   ISBN = {978-3-642-15759-2},
   ISSN = {0302-9743},
   language = {english},
   url = {http://www.fit.vutbr.cz/research/view_pub.php?id=9322}
}

Your IPv4 address: 23.20.141.48
Switch to IPv6 connection

DNSSEC [dnssec]