Automatic speech segmentation with the application of the Czech TTS system

0303740 - URE-Y 20000108 RIV DE eng C - Konferenční příspěvek (zahraniční konf.)
Horák, Petr - Hesounová, Alžběta
Automatic speech segmentation with the application of the Czech TTS system.
Berlin: Springer, 2000. Lecture Notes in Artificial Intelligence., 1902. ISBN 3-540-41042-2. In: Text, Speech and Dialogue. - (Sojka, P.; Kopeček, I.; Pala, K.), s. 201-206
[TSD /3./. Brno (CZ), 13.05.2000-16.05.2000]
Grant CEP: GA ČR GV102/96/K087
Grant ostatní: EU COST(XE) OC 258.10
Výzkumný záměr: CEZ:AV0Z2067918
Klíčová slova: speech synthesis * speech processing
Kód oboru RIV: BD - Teorie informace

This article presents automatic phonetic segmentation of natural speech based on the use of a speech synthesizer and dynamic time warping (DTW) algorithm. The speech synthesizer is used to create a synthetic reference speech pattern with phonetic segmentation infomation (phonemes, diphones, syllables, intonation units etc.). The reference synthetic speech pattern is then used in the alignment process. The main motivation for this work lay in the lack of usable segmentation tools for Czech.
Trvalý link: http://hdl.handle.net/11104/0113927