# X-LANCE/AniTalker

[ACM MM 2024] This is the official code for "AniTalker: Animate Vivid and Diverse Talking Faces through Identity-Decoupled Facial Motion Encoding"

Repository: https://github.com/X-LANCE/AniTalker
Canonical: https://ross.abutalabs.com/products/anitalker
Homepage: https://x-lance.github.io/AniTalker/
Language: Jupyter Notebook
License: Apache-2.0
License Family: permissive
Last push: 2024-08-15T15:42:24+00:00

## Health v2 (maintenance only)
Score: 24/100 (v2, computed 2026-09-03T02:20:16.233290+00:00)
- activity 0, release rhythm 35, longevity 61
- inputs: {"age_days": 857, "days_push": 748, "days_rel": null, "gap_med": null, "n_releases_24m": 0}
- flags: no_releases
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 1598, forks 144 (observed 2026-08-28T04:05:09.324992+00:00)

## What it is
AniTalker is the official PyTorch implementation of an ACM MM 2024 paper that animates a single static portrait into a vivid talking-face video driven by input audio. It uses identity-decoupled facial motion encoding with self-supervised learning and a diffusion model with a variance adapter to produce diverse, realistic facial expressions and head movements.

## Use cases
- generate a talking head video from a single portrait and audio clip
- animate an avatar's face with lip sync and nonverbal expressions
- create diverse talking face animations from the same control signals
- build dynamic digital avatars for virtual presenters
- research identity-decoupled facial motion representation
- run a web UI or Colab demo for audio-driven face animation

## When to choose
- you need audio-driven talking face generation from a single image
- you want diverse, nonverbal facial motion rather than only lip sync
- you are reproducing or extending the AniTalker research
- you want a ready web UI, Colab, or Hugging Face demo to try the model

## When to avoid
- you need real-time or production-grade streaming avatar generation
- you lack a GPU or cannot download external checkpoints
- you need fine-grained manual control over facial rigging like 3D puppet tools
- you need a maintained product with long-term support rather than research code

## Facets
- artifact type: library
- maturity: active
- function: machine-learning, deep-learning, image-processing, audio-processing, video-processing
- domain: artificial-intelligence, computer-vision, deep-learning, media
- platform: python, windows
- tags: talking-face, talking-head, face-animation, audio-driven, diffusion-model, avatar-generation, lip-sync, research-code, acm-mm-2024, linux, macos, gpu

## Member repositories
- X-LANCE/AniTalker (main) score 24

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:05:09.324992+00:00.
- Health v2: computed from the inputs above; adoption is never an input.
- Inferred fields (summary, facets, guidance): AI-extracted, prompt v1, taxonomy v1, on 2026-08-30T03:53:15.781959+00:00, confidence not recorded.
  - readme: https://github.com/X-LANCE/AniTalker (fetched 2026-08-28T04:05:09.324992+00:00, sha 7e6a509cc6aa)
  - homepage: https://x-lance.github.io/AniTalker/ (fetched 2026-08-29T11:24:35.748866+00:00, sha 5fe6200d37e4)
- Data as of 2026-08-30T08:39:29.467469+00:00.