Publication Details

TRAP-based Techniques for Recognition of Noisy Speech

GRÉZL, F.; ČERNOCKÝ, J. TRAP-based Techniques for Recognition of Noisy Speech. In Proc. 10th International Conference on Text Speech and Dialogue (TSD 2007). LNCS. Berlin: Springer Verlag, 2007. p. 270-277. ISBN: 978-3-540-74627-0.
Czech title
Techniky založené na TRAP pro rozpoznávání zašuměné řeči
Type
conference paper
Language
English
Authors
Grézl František, Ing., Ph.D. (DCGM)
Černocký Jan, prof. Dr. Ing. (DCGM)
URL
Keywords

TRAP techniques, noisy speech recognition, multistream processing, feature combination

Abstract

Paper persents performance of TRAP tecniques on noisy speech data, showing improvement using separate information processing with following combination of processed streams.

Annotation

This paper presents a systematic study of performance of TempoRAl Patterns (TRAP) based features and their proposed modifications and combinations for speech recognition in noisy environment. The experimental results are obtained on AURORA2 database with clean training data. We observed large dependency of performance of different TRAP modifications on noise level. Earlier proposed TRAP system modifications help in clean conditions but degrade the system performance in presence of noise. The combination techniques on the other hand can bring large improvement in case of weak noise and degrade only slightly for strong noise cases. The vector concatenation combination technique is improving the system performance up to strong noise.

Published
2007
Pages
270–277
Proceedings
Proc. 10th International Conference on Text Speech and Dialogue (TSD 2007)
Series
LNCS
ISBN
978-3-540-74627-0
Publisher
Springer Verlag
Place
Berlin
BibTeX 
@inproceedings{BUT30571,
  author="František {Grézl} and Jan {Černocký}",
  title="TRAP-based Techniques for Recognition of Noisy Speech",
  booktitle="Proc. 10th International Conference on Text Speech and Dialogue (TSD 2007)",
  year="2007",
  series="LNCS",
  pages="270--277",
  publisher="Springer Verlag",
  address="Berlin",
  isbn="978-3-540-74627-0"
}
Back to top