Synthesizing Prosody for Commands in a Xhosa TTS System

摘要

Xhosa is an Afican tone language spoken in South Africa. The relationship between the prosodic features duration, pitch and loudness of Xhosa commands was determined through acoustic analysis and perceptual experimetnation. Combining the resutls of the acoustic analysis and the perceptual experiemtn proved to be an appropriate method of parameter extraction with which to formulate a prosodic model for the generation of perceptually acceptable imperatives in a practical Xhosa TTS system.