Skip to content

Publications

Detail of publication

Citation

Jan Romportl : Automatic Prosodic Phrase Annotation in a Corpus for Speech Synthesis . Proceedings of Speech Prosody 2010, University of Illionois, Chicago, IL, USA, 2010.

Download PDF

PDF

Abstract

In order to improve speech naturalness of a unit selection TTS system it is necessary to annotate prosodic phrase boundaries in the whole source corpus, which is extremely difficult to achieve manually. It is thus usefull to employ a machine classifier. This paper discusses suitable feature selection for such classification of a Czech TTS corpus, presents results of experiments with linear and quadratic classifiers and artificial neural networks, and compares them with human annotators.

Detail of publication

Title: Automatic Prosodic Phrase Annotation in a Corpus for Speech Synthesis
Author: Jan Romportl
Language: English
Date of publication: 11 May 2010
Year: 2010
Type of publication: Papers in proceedings of reviewed conferences
Book title: Proceedings of Speech Prosody 2010
Publisher: University of Illionois
Address: Chicago, IL, USA
Date: 11 May 2010 - 14 May 2010
/ 2010-05-28 15:25:16 /

Keywords

speech synthesis, prosody, prosodic phrase, classification, neural network, unit selection, corpus

BibTeX

@INPROCEEDINGS{JanRomportl_2010_AutomaticProsodic,
 author = {Jan Romportl},
 title = {Automatic Prosodic Phrase Annotation in a Corpus for Speech Synthesis},
 year = {2010},
 publisher = {University of Illionois},
 address = {Chicago, IL, USA},
 booktitle = {Proceedings of Speech Prosody 2010},
 url = {http://www.kky.zcu.cz/en/publications/JanRomportl_2010_AutomaticProsodic},
}