Result Details

Recognition of Phoneme Strings using TRAP Technique

SCHWARZ, P.; MATĚJKA, P.; ČERNOCKÝ, J. Recognition of Phoneme Strings using TRAP Technique. Proceedings of 8th International Conference Eurospeech. European Conference EUROSPEECH. Geneve: International Speech Communication Association, 2003. no. 9, p. 1-4. ISSN: 1018-4074.
Type
conference paper
Language
English
Authors
Schwarz Petr, Ing., Ph.D.
Matějka Pavel, Ing., Ph.D.
Černocký Jan, prof. Dr. Ing.
Abstract

We investigate and compare several techniques for automatic recognitionof unconstrained context-independent phoneme strings from TIMIT andNTIMIT databases. Among the compared techniques, the technique based onTempoRAl Patterns (TRAP) achieves the best results in the clean speech,it achieves about 10% relative improovements against baseline system.Its advantage is also observed in the presence of mismatch betweentraining and testing conditions. Issues such as the optimal length oftemporal patterns in the TRAP technique and the effectiveness of meanand variance normalization of the patterns and the multi-band input theTRAP estimations, are also explored.

Keywords

speech recognition, feature extraction, TRAP

URL
Published
2003
Pages
1–4
Journal
European Conference EUROSPEECH, vol. 2003, no. 9, ISSN 1018-4074
Proceedings
Proceedings of 8th International Conference Eurospeech
Conference
Eurospeech 2003-Switzerland - 8th European conference on speech communication and technology
Publisher
International Speech Communication Association
Place
Geneve
BibTeX 
@inproceedings{BUT14174,
  author="Petr {Schwarz} and Pavel {Matějka} and Jan {Černocký}",
  title="Recognition of Phoneme Strings using TRAP Technique",
  booktitle="Proceedings of 8th International Conference Eurospeech",
  year="2003",
  journal="European Conference EUROSPEECH",
  volume="2003",
  number="9",
  pages="1--4",
  publisher="International Speech Communication Association",
  address="Geneve",
  issn="1018-4074",
  url="http://www.fit.vutbr.cz/~schwarzp/publi/2003/euro2003.pdf"
}
Projects
Data driven and anthropic coding and recognition of speech, GACR, Postdoktorandské granty, GP102/02/D108, start: 2002-09-01, end: 2005-08-30, completed
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 Radio Electronics (UREL)
Faculty of Electrical Engineering and Communication (FEEC)
Faculty of Information Technology (FIT)
Back to top