Skip to content

Publications

Detail of publication

Citation

Matoušek, J. and Tihelka, D. and Psutka, J. : Automatic segmentation for Czech concatenative speech synthesis using statistical approach with boundary-specific correction . Eurospeech 2003 - Interspeech, proceedings of the 8th European Conference on Speech Communication and Technology, p. 301-304, ISCA, Geneva, Switzerland, 2003.

Download PDF

PDF

Abstract

This paper deals with the problems of automatic segmentation for the purposes of Czech concatenative speech synthesis. Statistical approach to speech segmentation using HMMs is applied in the baseline system. Several improvements of this system are then proposed to get more accurate segmentation results. These enhancements mainly concern the various strategies of HMM initialization (flat-start initialization, hand-labeled or speaker independent HMM bootstrapping). Since HTK was utilized in our work, a correction of the output boundary placements is proposed to reflect speech parameterization mechanism. An objective comparison of various automatic methods and manual segmentation is performed to find out the best method. The best results were obtained for boundary-specific statistical correction of the segmentation that resulted from bootstrapping with hand-labeled HMMs (96% segmentation accuracy in tolerance region 20ms).

Detail of publication

Title: Automatic segmentation for Czech concatenative speech synthesis using statistical approach with boundary-specific correction
Author: Matoušek, J. ; Tihelka, D. ; Psutka, J.
Language: English
Date of publication: 1 Sep 2003
Year: 2003
Type of publication: Papers in proceedings of reviewed conferences
Book title: Eurospeech 2003 - Interspeech, proceedings of the 8th European Conference on Speech Communication and Technology
Page: 301 - 304
ISSN: 1018-4074
Publisher: ISCA
Address: Geneva, Switzerland
Date: 1 Sep 2003 - 4 Sep 2003
/ 2008-05-20 11:47:16 /

Keywords

HMM-based speech segmentation, shifting HTK boundaries, HMM initialization, flat-start, bootstrap, boundary-specific statistical correction, objective comparison, concatenative speech synthesis

BibTeX

@INPROCEEDINGS{MatousekJ_2003_Automatic,
 author = {Matou\v{s}ek, J. and Tihelka, D. and Psutka, J.},
 title = {Automatic segmentation for Czech concatenative speech synthesis using statistical approach with boundary-specific correction},
 year = {2003},
 publisher = {ISCA},
 address = {Geneva, Switzerland},
 pages = {301-304},
 booktitle = {Eurospeech 2003 - Interspeech, proceedings of the 8th European Conference on Speech Communication and Technology},
 ISSN = {1018-4074},
 url = {http://www.kky.zcu.cz/en/publications/MatousekJ_2003_Automatic},
}