Description
Lip sync and expressions for avatars or game characters mean frame-by-frame work or expensive motion capture. NVIDIA's Audio2Face-3D generates high-fidelity 3D facial animation from audio, from recorded files or real-time streams.
It analyzes speech to synchronize jaw, tongue and eye motion, and this repo collects the models and tools.
Audio-driven:Animation from speech.
Real time:Live audio streams.
Detail:Jaw, tongue and eyes.
Toolkit:Official models and tools.
It analyzes speech to synchronize jaw, tongue and eye motion, and this repo collects the models and tools.
Features
Audio-driven:Animation from speech.
Real time:Live audio streams.
Detail:Jaw, tongue and eyes.
Toolkit:Official models and tools.
Screenshots
Tags:animation