audio2face-3d Model by NVIDIA | NVIDIA NIM

Converts streamed audio to facial blendshapes for realtime lipsyncing and facial performances.