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
Demo url: https://jetfontanilla.github.io/canvas-talking-head-model/
For generating Viseme JSON Data: https://github.com/jetfontanilla/azure-viseme-json