Přejít na obsah

Detail publikace


Santarius, J. and Tihelka, D. : Prosody modeling of synthetic speech . ECMS 2003, p. 89-92, Technical University, Liberec, 2003.


One of the limitations of today’s TTS systems is lack of prosody. Prosody parameters, such as pitch period, duration and intensity, are mostly lived monotone (un-generated) or are generated according to the simply rules. However prosody parameters influence comprehensibility and naturalness of synthetic speech. There was created a prosody synthesizer on CTU Prague, to be able to implement new algorithms of prosody modelling. However created prosody synthesizer had not disposed by suitable TTS system. Therefore cooperation was made between CTU Prague and UWB Pilsen. A special user interface was created to the UWB’s TTS system ARTIC. Established Internet interface enables full control of prosody parameters such as melody, duration and intensity of synthesized speech. More information about this project can be found on Internet address: http://amber.feld.cvut.cz/user/prosody.

Detail publikace

Název: Prosody modeling of synthetic speech
Autor: Santarius, J. ; Tihelka, D.
Jazyk publikace: anglicky
Datum vydání: 2.6.2003
Rok vydání: 2003
Typ publikace: Stať ve sborníku
Název časopisu / knihy: ECMS 2003
Strana: 89 - 92
ISBN: 80-7083-708-X
Nakladatel: Technical University
Místo vydání: Liberec
Datum: 2.6.2003 - 4.6.2003
/ /

Klíčová slova

prosody of speech, text-to-speech system, continued modulation, fujisaki's model, neural networks, ARTIC


 author = {Santarius, J. and Tihelka, D.},
 title = {Prosody modeling of synthetic speech},
 year = {2003},
 publisher = {Technical University},
 journal = {ECMS 2003},
 address = {Liberec},
 pages = {89-92},
 ISBN = {80-7083-708-X},
 url = {http://www.kky.zcu.cz/en/publications/SantariusJ_2003_Prosodymodelingof},