Intonation modeling for Croatian speech synthesis

Intonation modeling has become a very important part of speech synthesis and speech recognition systems. This paper summarizes basic concepts of prosody. It outlines prosody modeling approaches, labeling systems and approaches to the generation of prosodic information with three different prosody models: ToBI, Tilt and Fujisaki. Goals and significance of research of Croatian prosody modeling and its implementation to TTS systems are discussed.