Click "Speak Text" to begin text-to-speech with lip sync
REST
Lip Sync Technology
Audio Processing: Web Speech API generates speech with onboundary events marking word boundaries. Each character maps to visemes (A,E,I,O,U) representing mouth shapes.
Sync Method: Real-time word boundary events trigger viseme sequences. Characters animate over ~60ms intervals with smooth interpolation between expressions.
Avatar Tech: VRM model with Three.js WebGL rendering. Expression morphs blend using weighted targets with 35ms smoothing for natural transitions.