A 3D communication platform based on text-to-visual speech sythesis

A 3D real-time communication platform based on text-to-visual speech (TTVS) synthesis is proposed. Partners of a communication pair exchange messages with words. At each end interface, TTVS technology is adopted to synthesize received textual messages into speeches. Simultaneously, a 3D avatar expresses embedded human emotions and intentions with body languages and facial expressions. Structure of such a communication platform is introduced. Expression models of human emotions and intentions are proposed. Autonomous interaction models of avatar whose master falls into operation silence are proposed. Driven by program, avatar´s movement is obtained by controlling joint angles along its freedom and their varying rates based on proposed methods. A prototype of 3D communication platform with function of TTVS is developed under VB6.0 and OpenGL.