# numz/sd-wav2lip-uhq

Wav2Lip UHQ extension for Automatic1111

Repository: https://github.com/numz/sd-wav2lip-uhq
Canonical: https://ross.abutalabs.com/products/sd-wav2lip-uhq
Language: Python
License: Apache-2.0
License Family: permissive
Topics: lip-sync, lipsync, stable-diffusion-web-ui, stable-diffusion-webui, stable-diffusion-webui-plugin, wav2lip, audio-driven-talking-face, deep-fake, deep-fakes, image-animation, talking-face, talking-face-generation, talking-head, meta-human-creator, face-swapping, faceswap
Last push: 2024-06-14T14:30:03+00:00

## Health v2 (maintenance only)
Score: 28/100 (v2, computed 2026-09-02T17:46:02.011165+00:00)
- activity 0, release rhythm 35, longevity 80
- inputs: {"age_days": 1126, "days_push": 810, "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 1424, forks 185 (observed 2026-08-28T04:04:41.351773+00:00)

## What it is
A Wav2Lip Studio extension for the Stable Diffusion WebUI (Automatic1111) that generates high-quality lip-synced talking-face videos from an input video and speech audio file. It enhances the base Wav2Lip output using Stable Diffusion-based post-processing, with optional face swap (facefusion), GFPGAN face restoration, and Bark text-to-speech integration.

## Use cases
- generate lip-synced video from audio
- make a talking head video from a photo or clip
- improve wav2lip video quality
- sync a character's mouth to speech audio
- create deepfake talking face videos
- add text-to-speech narration to an avatar video
- swap faces in a lip-synced video

## When to choose
- you already use Automatic1111's Stable Diffusion WebUI and want lip-sync generation inside it
- you need higher-quality lip-sync than vanilla Wav2Lip produces
- you want an all-in-one pipeline with TTS, face swap, and face restoration

## When to avoid
- you need a standalone CLI or API without the Automatic1111 WebUI
- you have no GPU or limited VRAM, since generation is slow and memory-heavy
- you need production-grade, multi-face or 4K lip-sync workflows (only in the paid standalone version)

## Facets
- artifact type: plugin
- maturity: active
- function: video-processing, speech-recognition, image-processing, machine-learning, tts
- domain: artificial-intelligence, media, deep-learning
- platform: python, cross-platform
- tags: wav2lip, lip-sync, talking-head, stable-diffusion-webui, automatic1111-extension, face-swap, deepfake, audio-driven-animation, gfpgan, bark, video, gpu

## Member repositories
- numz/sd-wav2lip-uhq (main) score 28

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:04:41.351773+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 2026-08-30T04:37:35.919052+00:00, confidence not recorded.
  - readme: https://github.com/numz/sd-wav2lip-uhq (fetched 2026-08-28T04:04:41.351773+00:00, sha 7f094fbd6ce7)
- Data as of 2026-08-30T08:39:29.467469+00:00.