Skip to content

Publications

Detail of publication

Citation

Psutka Josef V. and Pražák Aleš : Multiple Application of the MLLT Based on Clustering Supported by Phonetic Knowledge . 9th International Conference on Signal Processing Proceedings, vol. ;, p. 613-617, IEEE, Peking, Čína, 2008.

Abstract

This paper intends to show the effect of multiple use of a Maximum Likelihood Linear Transform (MLLT) in continuous speech recognition tasks. The correspon- ding covariance matrices that were subjected to the transform were clustered using phonetic knowledge. The impact of this approach was tested on the ASRs with PLP- and MFCC-based front-ends and was compared with other known feature decorrelation and dimension reduction techniques.

Detail of publication

Title: Multiple Application of the MLLT Based on Clustering Supported by Phonetic Knowledge
Author: Psutka Josef V. ; Pražák Aleš
Language: English
Year: 2008
Type of publication: Papers in proceedings of reviewed conferences
Title of journal or book: 9th International Conference on Signal Processing Proceedings
Číslo vydání: ;
Page: 613 - 617
ISBN: 978-1-4244-2178-7
Publisher: IEEE
Address: Peking, Čína
Date: 29 Oct 2008
2011-03-15 16:21:32 / 2011-06-09 12:30:28 / 1

Keywords

multiple MLLT, decorrelation, speech recognition

BibTeX

@INPROCEEDINGS{PsutkaJosefV_2008_MultipleApplication,
 author = {Psutka Josef V. and Pra\v{z}\'{a}k Ale\v{s}},
 title = {Multiple Application of the MLLT Based on Clustering Supported by Phonetic Knowledge},
 year = {2008},
 publisher = {IEEE},
 journal = {9th International Conference on Signal Processing Proceedings},
 address = {Peking, \v{C}\'{i}na},
 volume = {;},
 pages = {613-617},
 ISBN = {978-1-4244-2178-7},
 note = {Dopln\v{e}no UT ISI},
 url = {http://www.kky.zcu.cz/en/publications/PsutkaJosefV_2008_MultipleApplication},
}