numz/sd-wav2lip-uhq

Health v2 · maintenance only

Flags: no_releases

Adoption not part of the score

1424 stars · 185 forks observed · 2026-08-28

What it is AI-extracted, prompt v1, taxonomy v1, 2026-08-30, confidence not recorded

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

plugin · maturity active

video-processing speech-recognition image-processing machine-learning tts artificial-intelligence media deep-learning python cross-platform wav2lip lip-sync talking-head stable-diffusion-webui automatic1111-extension face-swap deepfake audio-driven-animation gfpgan bark video gpu

Member repositories

For agents

markdown · JSON · MCP: product_card(name="numz/sd-wav2lip-uhq")