LUCIA a new emotive/expressive Italian talking head

In this chapter we present a study on audiovisual speech analysis and synthesis, for the creation of natural and expressive talking heads. The framework for the modeling of facial and lip movements, and the visual design of an animated MPEG-4 talking face, are first discussed. Then, the design of the voice synthesis modules aimed at producing expressive/emotive speech synthesis, is described. The resulting audiovisual synthesis system integrates voice synthesis, facial animation, and a markup language based interface, that allows to produce synchronized and emotionally coherent audio and video outputs from tagged text.