Botinis et al. (2001)

Let us discuss and summarize the following issues.

(1) Which linguistic and paralinguistic functions are subserved by the
intonation of utterances?

(2) What is the role of intonation models in text-to-speech (TTS)
synthesis systems? What kind of input information does the intonation
model need and which other TTS components can it inform?

(3) To what extent is intonational structure exploited in automatic
speech recognition (ASR) systems? Can the detection of intonation
events improve overall ASR performance?