Forked from jetfontanilla
Proof of concept for a canvas-based talking head model
This uses the following technologies:
- ChatGPT for generating script
- Azure Text-to-speech for generating audio
- Viseme for generating the phoneme frame data
- Canvas for drawing the mouth animations