ISSN 0021-3454 (print version)
ISSN 2500-0381 (online version)
Menu

2
Issue
vol 67 / February, 2024
Article
UDC 519.688

A HYBRID TECHNOLOGY FOR TTS SYSTEM BASED ON HIDDEN MARKOV MODELS AND UNIT SELECTION ALGORITHM

P. . Chistikov
«Speech Technologies Center», Ltd.; scientific researcher


E. A. Korolkov
STC Ltd., St. Petersburg; Scientist


A. O. Talanov
STC Ltd., St. Petersburg, Speech Synthesis Department; Head of the Department


A. I. Solomennik
Speech Technology Ltd., Minsk; Scientist


Read the full article 

Abstract.   An approach to synthesis of Russian TTS system based on integration of Hidden Markov Models and Unit Selection algorithms is presented. The voice model creation method is developed for constructing a natural intonation contour. Improved quality of synthesized speech is confirmed by experimental results.
Keywords: speech synthesis, Hidden Markov Models, Unit Selection, voice model