Skip to content

ComfyUI-IF_MemoAvatar

Expressive talking-head video from one image and audio

Visit official site

What ComfyUI-IF_MemoAvatar is

The model is a memory-guided diffusion system that turns a single image and audio into an expressive talking-head video.

What you can do with it

  • Animate a face from one image
  • Drive facial motion with audio
  • Transfer emotional expression

Getting started

  • Clone the repository locally
  • Install Python and PyTorch
  • Add extra dependencies if needed
  • Place model files in the given directory

Pros & cons

✓ What we like

  • Custom model files and parameter tuning are supported
  • Linux and Windows are both named as supported

! What to watch out for

  • Python and PyTorch dependencies must be installed
  • Model files go into a specified directory

FAQ

What are the inputs?

The material says a single image and an audio input are used.

Where was it published?

The material says a related paper was released on arXiv.

Last reviewed: 2026-11-15

More AI video generation tools

View all →

How we review