Publication Details

Feature gaussianization in speech recognition

MATĚJKA, P.; ČERNOCKÝ, J. Feature gaussianization in speech recognition. Proc. 12th International scientific conference Radioelektronika 2002. Bratislava: Slovak University of Technology in Bratislava, 2002. p. 0-0. ISBN: 80-227-1700-2.
Czech title
Gaussianizace dat v rozpoznávání řeči
Type
conference paper
Language
English
Authors
Matějka Pavel, Ing., Ph.D. (DCGM)
Černocký Jan, prof. Dr. Ing. (DCGM)
URL
Keywords

speech recognition, Gaussianization, non-linear transform

Abstract

Feature gaussianization in speech recognition

Annotation

In Hidden Markov models, speech data are modeled by Gaussian distributions. In this paper, we propose to Gaussianize the features to better fit to this modeling. A distribution of the data is estimated and a transform function is derived. We test three methods of the transform estimation (global, speaker based, frame based) and report results on the SPINE 2000 task with SPhinx recognizer. We conclude that the proposed method is a cheap way to increase the recognition accuracy.

Published
2002
Pages
4
Proceedings
Proc. 12th International scientific conference Radioelektronika 2002
ISBN
80-227-1700-2
Publisher
Slovak University of Technology in Bratislava
Place
Bratislava
BibTeX 
@inproceedings{BUT10283,
  author="Pavel {Matějka} and Jan {Černocký}",
  title="Feature gaussianization in speech recognition",
  booktitle="Proc. 12th International scientific conference Radioelektronika 2002",
  year="2002",
  pages="4",
  publisher="Slovak University of Technology in Bratislava",
  address="Bratislava",
  isbn="80-227-1700-2",
  url="http://www.fit.vutbr.cz/~matejkap/publi/2002/RA2002.pdf"
}
Back to top