Expressive/Affective Speech Synthesis

Abstract

The focus of speech synthesis research has recently shifted from read speech towards more conversational styles of speech, in order to reproduce those situations where a speech synthesis is used as part of a dialogue. When a speech synthesizer speechsynthesizer is used to represent the voice of a cognisant agent, whether human or simulated, there is need for more than just the intelligible portrayal of linguistic information; there is also a need for the expression of affect. This chapter reviews some recent advances in the synthesis of expressive speech and shows how the technology can be adapted to include the display of affect in conversational speech. The chapter discusses how the presence of an interactive and active partner in a conversation can greatly affect the styles of human speech and presents a model of the cognitive processes that result in these differences, which concern not just the acoustic prosody and phonation quality of an utterance, but also its lexical selection and phrasing. It proposes a measure of the ratio of paralinguistic to linguistic content in an utterance as a means of quantifying the expressivity of a speaking style, and closes with a description of a phrase-level concatenative speech synthesis system that is currently in development.