metaai·lightalo unofficial · independent
Universe / Reality Labs Research / audio2photoreal
Reality Labs Research · 2024

audio2photoreal

Record your voice, watch a photoreal avatar gesture along

open source
★ 2.8k audio2photoreal❝ 99 citationsread 2026-09-03

Latest: CVPR 2024 (arXiv Jan 2024); code, dataset and Gradio demo live

Open code and dataset for synthesizing full-bodied photorealistic Codec Avatars — face, body, and hands — that gesture naturally from conversational audio alone. Combines vector-quantization sample diversity with diffusion for high-frequency motion detail. Released January 2024 by Meta Reality Labs researchers with UC Berkeley; published at CVPR 2024.

Why it matters

Synthesizes full-body photorealistic avatars, including face and hands, that gesture naturally from conversational audio alone, combining vector-quantized sample diversity with diffusion for high-frequency motion. Published at CVPR 2024 with code, a four-participant conversational dataset and a Colab demo under CC-BY-NC 4.0, it is a public window into Codec Avatar animation.

Facts

Try it yourself

Lineage

Descends fromCodec Avatars

See the whole family tree →

Sources

More in Reality Labs Research

OrionCodec Avatar Studioemg2qwertyemg2poseGeneric Neuromotor InterfaceMeta Ray-Ban Display + Neural Band

Read the Reality Labs Research story on the sky →

✦ Open on the map Explore Reality Labs Research Quiz me