lipsync

To lip-sync a face to an audio track, this Skill routes your inputs across RunComfy endpoints like OmniHuman, Sync Labs, Kling, and Creatify. It validates your inputs and issues the runcomfy run command with the correct model path and payload to generate natural mouth motion.

Yes, you can generate avatar-style mouth movements from a still portrait. The Skill selects the appropriate lip-sync model based on your input shape and quality requirements, coordinating the endpoint selection to animate a static face with natural mouth motion synced to audio.

The best way to lip-sync an existing video is to use the Skill to route your inputs to the appropriate RunComfy endpoint. It matches your intent for mouth-swap on video, validates the inputs, and executes the documented runcomfy run payload for consistent results.

Yes, this Skill supports generate-and-sync from a script. It coordinates endpoint selection across RunComfy APIs to generate video content from your script and applies lip-syncing to match the resulting avatar mouth movements to the generated audio.

Yes, you need the RunComfy CLI installed to perform lip-syncing. The Skill provides a consistent, repeatable invocation by using the documented runcomfy run payload, requiring the CLI to issue commands with the correct model path and input payload.