NVIDIA Releases Audio2Face-3D for Audio-Driven Facial Animation

tom_doerr · x · 2026-08-23

NVIDIA has released Audio2Face-3D, technology that generates high-fidelity 3D facial animation from an audio source, supporting both pre-recorded files and real-time streams.

Core Capabilities:

Toolkit:

The hub provides pre-trained models, a development SDK, plugins for Autodesk Maya and Unreal Engine 5, and sample datasets.

Original post →

More from Multimodal

Multimodal channel →