Result Details

Local Time-Frequency Operators in TRAPs For Speech Recognition

GRÉZL, F. Local Time-Frequency Operators in TRAPs For Speech Recognition. 6th International Conference, TSD 2003 České Budějovice, Czech Republic, September 2003 Proceedings. Lecture Notes in Computer Science. České Budějovice: University of West Bohemia in Pilsen, 2003. no. 9, p. 269-274. ISBN: 3-540-20024-X. ISSN: 0302-9743.
Type
conference paper
Language
English
Authors
Grézl František, Ing., Ph.D., FIT (FIT)
Abstract

Publication compares different local operators working in specral domain for speech recognition with the use of temporal trajectories (TRAP) technique. The evaluation is done on digits recognition task.

Keywords

speech, speech processing, feature extraction ,speech recognittion, TRAP, TRAP modifications

URL
Annotation

This paper describes tests with local operators applied to
critical-band speech
spectrum prior to  temporal patterns (TRAPs) feature extraction. Tests
are performed with an  HMM recognizer on connected-digits task. We show
that frequency differentiation, in combination with any other
operator, improves the word-error rate (WER) of TRAP-based
recognizer.

Published
2003
Pages
269–274
Journal
Lecture Notes in Computer Science, vol. 2003, no. 9, ISSN 0302-9743
Proceedings
6th International Conference, TSD 2003 České Budějovice, Czech Republic, September 2003 Proceedings
Conference
International Conference on Text Speech and Dialogue, TSD 2003
ISBN
3-540-20024-X
Publisher
University of West Bohemia in Pilsen
Place
České Budějovice
BibTeX 
@inproceedings{BUT14204,
  author="František {Grézl}",
  title="Local Time-Frequency Operators in TRAPs For Speech Recognition",
  booktitle="6th International Conference, TSD 2003 České Budějovice, Czech Republic, September 2003 Proceedings",
  year="2003",
  journal="Lecture Notes in Computer Science",
  volume="2003",
  number="9",
  pages="269--274",
  publisher="University of West Bohemia in Pilsen",
  address="České Budějovice",
  isbn="3-540-20024-X",
  issn="0302-9743",
  url="https://www.fit.vut.cz/research/publication/7282/"
}
Projects
Voice technologies for support of information society, GACR, Standardní projekty, GA102/02/0124, start: 2002-01-01, end: 2004-12-31, completed
Research groups
Speech Data Mining Research Group BUT Speech@FIT (RG SPEECH)
Departments
Department of Computer Graphics and Multimedia (DCGM)
Back to top