Publication Details

Parallel Training of Neural Networks for Speech Recognition

VESELÝ Karel, BURGET Lukáš and GRÉZL František. Parallel Training of Neural Networks for Speech Recognition. In: Prof. Text, Speech and Dialogue 2010. LNAI 6231, vol. 2010. Brno: Springer Verlag, 2010, pp. 439-446. ISBN 978-3-642-15759-2. ISSN 0302-9743.
Czech title
Paralelní trénování neuronových sítí pro rozpoznávání řeči
Type
conference paper
Language
english
Authors
Veselý Karel, Ing., Ph.D. (DCGM FIT BUT)
Burget Lukáš, doc. Ing., Ph.D. (DCGM FIT BUT)
Grézl František, Ing., Ph.D. (DCGM FIT BUT)
URL
Keywords

neural network, phoneme classification, posterior features, backpropagation training, data parallelization

Abstract

The paper is on Parallel Training of Neural Networks for Speech Recognition. A new parallel-training tool TNet was designed and optimized for multiprocessor computers. The training acceleration rates are reported on a phoneme-state classification task.

Annotation

The feed-forward multi-layer neural networks have significant importance in speech recognition. A new parallel-training tool TNet was designed and optimized for multiprocessor computers. The training acceleration rates are reported on a phoneme-state classification task.

Published
2010
Pages
439-446
Journal
Lecture Notes in Computer Science, vol. 2010, no. 9, ISSN 0302-9743
Proceedings
Prof. Text, Speech and Dialogue 2010
Series
LNAI 6231
Conference
13th International Conference on Text, Speech and Dialogue, TSD 2010, Brno, CZ
ISBN
978-3-642-15759-2
Publisher
Springer Verlag
Place
Brno, CZ
BibTeX 
@INPROCEEDINGS{FITPUB9319,
   author = "Karel Vesel\'{y} and Luk\'{a}\v{s} Burget and Franti\v{s}ek Gr\'{e}zl",
   title = "Parallel Training of Neural Networks for Speech Recognition",
   pages = "439--446",
   booktitle = "Prof. Text, Speech and Dialogue 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 = "https://www.fit.vut.cz/research/publication/9319"
}
Back to top