skills.rest
agentspace-so
agentspace-soOFFICIAL
Explore all skills by this author →
GitHub

lipsync

Generate lip-synced videos from audio or scripts via RunComfy CLI.

31|9|Updated Apr 30, 2026
Search Tags:#avatar#lipsync#video dubbing#audio-driven animation#runcomfy cli#mouth swap#sync labs
One-click install
npx skills add https://github.com/agentspace-so/runcomfy-agent-skills --skill lipsync
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: lipsync
Source: https://github.com/agentspace-so/runcomfy-agent-skills/tree/main/lipsync
Command: npx skills add https://github.com/agentspace-so/runcomfy-agent-skills --skill lipsync

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

It removes the manual, time-consuming work of making a face match spoken audio by automatically producing a lip-synced video from the right input type.

Core Features & Use Cases

  • Route selection across RunComfy lip-sync endpoints: Chooses the best provider and model based on whether you have a portrait still, an existing source video, or only a script.
  • Audio-to-video mouth driving: Syncs a provided audio track onto an existing video or generates an avatar-style talking-head from a portrait.
  • Script-to-speech lipsync: Supports workflows where you provide text but no pre-recorded audio, generating speech in-pass and syncing it.
  • Real-world example: Dub a product launch by keeping the original video and swapping in a translated voiceover so the mouth matches the new language.

Quick Start

Ask for a lipsynced dub of an existing video by providing the source video URL and the voiceover audio URL.

Frequently Asked Questions about lipsync

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I automatically sync an audio track to an existing video for dubbing?▼

Audio-driven lip sync automatically matches a face's mouth motion to a new audio track. This process swaps the original mouth movement to match a translated voiceover, producing a lip-synced video without manual animation.

Can I generate a talking-head video from a portrait still and a script?▼

Yes, script-to-speech lipsync generates speech in-pass from provided text and syncs it onto a portrait-still avatar. This workflow creates a lifelike talking-head video when you only have a script and no pre-recorded audio.

What's the best way to choose a lip sync model for different video inputs?▼

RunComfy CLI routes your input to the best provider based on your source material. It selects between OmniHuman, Sync Labs, Kling, and Creatify endpoints to handle portrait stills, source-video mouth swaps, or script-only generation workflows.

Do I need pre-recorded audio to perform a video mouth swap?▼

No, pre-recorded audio is not required for video mouth swaps. You can use a script-only generation workflow where the system generates speech in-pass from text and synchronizes the lip movements automatically.

How does YAML routing work for audio-driven animation across different providers?▼

YAML routing defines the path across OmniHuman, Sync Labs sync v2/Pro, Kling, and Creatify for audio-driven animation. It configures the correct JSON input schema and output directory before invoking the RunComfy CLI to execute the lip sync process.

Related Skills

Explore More Design & Creative Skills→
agentspace-soagentspace-so
31

ace-step

Generate, inpaint, and outpaint music tracks via RunComfy ACE Step endpoints.

Official
Intermediate
eachlabseachlabs
28

video-localization

Translate video audio, generate subtitles, and dub with voice cloning and lip sync.

Official
Advanced
doany-aidoany-ai
5

lipsync

Route lip-sync requests across RunComfy endpoints and issue runcomfy run commands.

Official
Advanced
SheshiyerSheshiyer
4

ai-avatar-video

Generate lip-synced avatar videos from an image and audio via inference.sh CLI.

Community
Intermediate