Skip to content

EMAGE

Co-speech gestures generated from audio

Visit official site

What emage is

The model is described as a unified holistic co-speech gesture model that turns audio into natural gesture motions through expressive masked audio gesture modeling.

What you can do with it

  • Animate a virtual assistant during an online meeting
  • Give a game character believable body motion
  • Attach gesture output to a voice assistant

Key capabilities

  • Body pose and gesture sequences are generated from audio
  • Speech and prosody cues guide the gesture modeling
  • Highly dynamic and expressive gestures are listed

Pros & cons

✓ What we like

  • The approach is described as a unified holistic model
  • Expressive masked audio gesture modeling is used

! What to watch out for

  • Audio input is needed before any gesture can be generated
  • The listed output is a sequence of poses and gestures

FAQ

What goes in?

The material states audio input is used to capture speech and prosody.

What comes out?

The material lists body pose and gesture action sequences.

Last reviewed: 2026-11-15

More AI video generation tools

View all →

How we review