Článek ve sborníku konference

GRÉZL František. The Role of Neural Network Size in TRAP/HATS Feature Extraction. In: Proceedings Text, Speech and Dialogue 2011. Plzeň: Springer Verlag, 2011, s. 315-322. ISBN 978-3-642-23537-5. ISSN 0302-9743.
Jazyk publikace:angličtina
Název publikace:The Role of Neural Network Size in TRAP/HATS Feature Extraction
Název (cs):Role velikosti neuronové sítě v extrakci příznaků pomocí TRAP/HATS
Strany:315-322
Sborník:Proceedings Text, Speech and Dialogue 2011
Konference:14th International Conference on Text, Speech and Dialogue
Řada knih:LNAI 6836
Místo vydání:Plzeň, CZ
Rok:2011
ISBN:978-3-642-23537-5
Časopis:Lecture Notes in Computer Science, roč. 2011, č. 9, DE
ISSN:0302-9743
Vydavatel:Springer Verlag
URL:http://www.fit.vutbr.cz/research/groups/speech/publi/2011/grezl_tsd2011.pdf [PDF]
Klíčová slova
Neural networks, feature extraction, probabilistic features
Anotace
Článek se zabývá účinností pravdepodobnostních příznaků ziskanych pomocí TRAP/HATS technik v systému pro rozpoznávání řeči. Velikosti neuronových sítí v obou úrovních procesu jsou měněny a vliv této změny je vyhodnocen.
Abstrakt
We study the role of sizes of neural networks (NNs) in TRAP (Tempo- RAl Patterns) and HATS (Hidden Activation TRAPS architecture) probabilistic features extraction. The question of sufficient size of band NNs is linked with the question whether the Merger is able to compensate for lower accuracy of band NNs. For both architectures, the performance increases with increasing size of Merger NN. For TRAP architecture, it was observed, that increasing band NN size over some value has not further positive effect on final performance. The situation is different when HATS architecture is employed - increasing size of band NNs has mostly negative effect on final performance. This is caused by merger not being able to efficiently exploit the information hidden in its input with increased size. The solution is proposed in form of bottle-neck NN which allows for arbitrary size output.
BibTeX:
@INPROCEEDINGS{
   author = {František Grézl},
   title = {The Role of Neural Network Size in TRAP/HATS Feature
	Extraction},
   pages = {315--322},
   booktitle = {Proceedings Text, Speech and Dialogue 2011},
   series = {LNAI 6836},
   journal = {Lecture Notes in Computer Science},
   volume = {2011},
   number = {9},
   year = {2011},
   location = {Plzeň, CZ},
   publisher = {Springer Verlag},
   ISBN = {978-3-642-23537-5},
   ISSN = {0302-9743},
   language = {english},
   url = {http://www.fit.vutbr.cz/research/view_pub.php.cs?id=9751}
}

Vaše IPv4 adresa: 54.89.60.200
Přepnout na IPv6 spojení

DNSSEC [dnssec]