Conference paper

HSIAO Roger, NG Tim, GRÉZL František, KARAKOS Damianos, TSAKALIDIS Stavros, NGUYEN Long and SCHWARTZ Richard. Discriminative Semi-supervised Training for Keyword Search in Low Resource Languages. In: Proceedings of ASRU 2013. Olomouc: IEEE Signal Processing Society, 2013, pp. 440-445. ISBN 978-1-4799-2755-5.
Publication language:english
Original title:Discriminative Semi-supervised Training for Keyword Search in Low Resource Languages
Title (cs):Diskriminativní částečně kontrolované trénování vyhledávání klíčových slov pro jazyky s omezenými zdroji
Pages:440-445
Proceedings:Proceedings of ASRU 2013
Conference:IEEE 2013 Workshop on Automatic Speech Recognition and Understanding
Place:Olomouc, CZ
Year:2013
ISBN:978-1-4799-2755-5
Publisher:IEEE Signal Processing Society
URL:http://www.fit.vutbr.cz/research/groups/speech/publi/2013/hsiao_asru2013_0000440.pdf [PDF]
Keywords
semi-supervised training, low resource languages, keyword spotting
Annotation
This article is about Discriminative Semi-supervised Training for Keyword Search in Low Resource Languages.
Abstract
In this paper, we investigate semi-supervised training for low resource languages where the initial systems may have high error rate ( 70.0% word eror rate). To handle the lack of data, we study semi-supervised techniques including data selection, data weighting, discriminative training and multilayer perceptron learning to improve system performance. The entire suite of semi-supervised methods presented in this paper was evaluated under the IARPA Babel program for the keyword spotting tasks. Our semi-supervised system had the best performance in the OpenKWS13 surprise language evaluation for the limited condition. In this paper, we describe our work on the Turkish and Vietnamese systems.
BibTeX:
@INPROCEEDINGS{
   author = {Roger Hsiao and Tim Ng and Franti{\v{s}}ek Gr{\'{e}}zl and
	Damianos Karakos and Stavros Tsakalidis and Long Nguyen and
	Richard Schwartz},
   title = {Discriminative Semi-supervised Training for Keyword Search
	in Low Resource Languages},
   pages = {440--445},
   booktitle = {Proceedings of ASRU 2013},
   year = {2013},
   location = {Olomouc, CZ},
   publisher = {IEEE Signal Processing Society},
   ISBN = {978-1-4799-2755-5},
   language = {english},
   url = {http://www.fit.vutbr.cz/research/view_pub.php?id=10508}
}

Your IPv4 address: 54.198.134.104
Switch to IPv6 connection

DNSSEC [dnssec]