# OpenTalker/SadTalker

[CVPR 2023] SadTalker：Learning Realistic 3D Motion Coefficients for Stylized Audio-Driven Single Image Talking Face Animation

Repository: https://github.com/OpenTalker/SadTalker
Canonical: https://ross.abutalabs.com/products/sadtalker
Homepage: https://sadtalker.github.io/
Language: Python
License: NOASSERTION
License Family: other
Topics: audio-driven-talking-face, cvpr2023, talking-head, deep-fake, deep-fakes, image-animation, talking-face, talking-face-generation, talking-heads
Last push: 2024-06-26T12:51:53+00:00

## Health v2 (maintenance only)
Score: 22/100 (v2, computed 2026-09-03T02:20:16.233290+00:00)
- activity 0, release rhythm 8, longevity 98
- inputs: {"age_days": 1380, "days_push": 798, "days_rel": null, "gap_med": null, "n_releases_24m": 0}
- flags: no_license
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 14040, forks 2691 (observed 2026-08-28T04:11:05.365235+00:00)

## What it is
SadTalker is a CVPR 2023 deep learning tool that generates realistic talking head videos from a single portrait image and an audio clip by predicting 3DMM motion coefficients (head pose, expression) from audio. It is available as a Python codebase with Gradio demos, Hugging Face Spaces, Colab, and Stable Diffusion WebUI integrations.

## Use cases
- animate a portrait photo with speech audio
- generate talking head video from a single image
- create lip-synced avatar videos from voice recordings
- make a digital avatar speak from text-to-speech audio
- research audio-driven face animation
- produce deepfake-style talking face videos

## When to choose
- you need to animate a single still portrait with an audio track
- you want a research-grade, widely cited talking-head generation model
- you want ready-made demos via Hugging Face, Colab, or Stable Diffusion WebUI
- you need controllable head pose and expression styles

## When to avoid
- you need real-time lip sync for live streaming or video calls
- you require full-body avatar animation rather than a single face
- you need a commercially supported product with SLAs
- you lack a GPU and cannot tolerate slow inference

## Facets
- artifact type: application
- maturity: maintenance
- function: deep-learning, video-processing, audio-processing, machine-learning, llm-inference
- domain: deep-learning, computer-vision, artificial-intelligence
- platform: python, cross-platform
- tags: talking-head, talking-face-generation, audio-driven-animation, deepfake, face-animation, cvpr2023, 3dmm, generative-ai, audio, video, gpu, docker

## Member repositories
- OpenTalker/SadTalker (main) score 22

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:11:05.365235+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 2026-08-29T17:12:51.567335+00:00, confidence not recorded.
  - readme: https://github.com/OpenTalker/SadTalker (fetched 2026-08-28T04:11:05.365235+00:00, sha a594a7558191)
  - homepage: https://sadtalker.github.io/ (fetched 2026-08-29T08:06:44.777490+00:00, sha 140f8d74879e)
- Data as of 2026-08-30T08:39:29.467469+00:00.