Faculty of Information Technology, BUT

Publication Details

Sub-word modeling of out of vocabulary words in spoken term detection

SZŐKE Igor, BURGET Lukáš, ČERNOCKÝ Jan and FAPŠO Michal. Sub-word modeling of out of vocabulary words in spoken term detection. In: Proc. 2008 IEEE Workshop on Spoken Language Technology. Goa: IEEE Signal Processing Society, 2008, p. 4. ISBN 978-1-4244-3472-5.
Czech title
Pod slovní modelování slov mimo slovník v detekci mluvených výrazů
Type
conference paper
Language
english
Authors
URL
Keywords
phone, multigram, spoken term detection, subword, keyword spotting, syllable, lattice

Abstract
The work is on sub-word modeling of out of vocabulary words in spoken term detection
Annotation
This paper deals with comparison of sub-word based methods for
spoken term detection (STD) task and phone recognition. The subword units are needed for search for out-of-vocabulary words. We compared words, phones and multigrams. The maximal length and pruning of multigrams were investigated first. Then two constrained methods of multigram training were proposed. We evaluated on the NIST STD06 dev-set CTS data. The conclusion is that the proposed method improves the phone accuracy more than 9% relative and STD accuracy more than 7% relative.

Published
2008
Pages
4
Proceedings
Proc. 2008 IEEE Workshop on Spoken Language Technology
Conference
2008 IEEE Workshop on Spoken Language Technology, Goa, IN
ISBN
978-1-4244-3472-5
Publisher
IEEE Signal Processing Society
Place
Goa, IN
BibTeX
@INPROCEEDINGS{FITPUB8840,
   author = "Igor Sz\H{o}ke and Luk\'{a}\v{s} Burget and Jan \v{C}ernock\'{y} and Michal Fap\v{s}o",
   title = "Sub-word modeling of out of vocabulary words in spoken term detection",
   pages = 4,
   booktitle = "Proc. 2008 IEEE Workshop on Spoken Language Technology",
   year = 2008,
   location = "Goa, IN",
   publisher = "IEEE Signal Processing Society",
   ISBN = "978-1-4244-3472-5",
   language = "english",
   url = "https://www.fit.vut.cz/research/publication/8840"
}
Back to top