Audio2Face-3D

Model by NVIDIA.

Modelspeech audio in, facial animation out

Input

speech audio

Output

facial animation

Speech audio to facial animation.

What you send: speech audio
What you get: facial animation

Available as a Pack. Write to [email protected].

NVIDIA's SDK runs its Audio2Face-3D v3.0 model on NVIDIA's TensorRT. We run the network natively on a TPU v5e chip, and it matched the reference output in 72 of 72 test cases (worst relative error 1.44e-4). Measured 23 October 2026.

More models

Track2Rig

Model Experimental by MASHBOX Labs

3D point tracks of a moving body in; skeleton, skin and bone motion out as GLB.

3D point tracks in, skeleton, skin and motion (GLB) out

MASHBOX Stride

Model by MASHBOX Labs

Describe a movement and get a humanoid motion clip.

text in, motion clip out

Z-Image Turbo

Model by Tongyi-MAI (Alibaba)

Text to image.

text in, image out

Wan 2.2 Image to Video

Model by Wan team (Alibaba)

An image and a prompt in, a video clip out.

image + text in, video out