Modelování prozodie v české text-to-speech syntéze

Článek popisuje datový přístup k modelování všech tří základních prozodických vlastností - základního hlasivkového tónu, intenzity a trvání.This paper describes data-driven modelling of all three basic prosodic features - fundamental frequency, intensity and segmental duration - in the Czech text-to-speech system ARTIC. The fundamental frequency is generated by a model based on concatenation of automatically acquired intonational patterns. Intensity of synthesised speech is modelled by experimentally created rules which are in conformity with phonetics studies. Phoneme duration modelling has not been previously solved in ARTIC and this paper presents the first solution to this problem using a CART-based approach