ugc-manual
by pauldelavallaz
Coding Agents & IDEsBrief
Generate lip-sync video from image + user's own audio recording. ✅ USE WHEN: - User provides their OWN audio file.
Description
What This Skill Does
Generates a lip-sync video by combining a face image with a user-provided audio recording. The person in the image is animated to match the audio using ComfyDeploy's UGC-MANUAL workflow. Use this when you already have an audio file and want to preserve the exact voice and timing.
Preserves the exact voice, timing, and delivery of the user's own recording instead of replacing it with AI-generated speech.
When to Use It
- Animating a profile photo to match a recorded voice message
- Turning a Telegram or WhatsApp voice note into a talking-head video
- Syncing a character image to externally generated TTS audio
- Creating a video avatar that speaks in your own voice
- Producing lip-synced content from pre-recorded audio clips
Example Workflow
User asks: Create a lip-sync video from my profile photo and this MP3 voice recording
- 1
Accept the image file and audio file from the user
- 2
Convert audio to WAV PCM 16-bit mono 48kHz via ffmpeg if not already in that format
- 3
Submit the image and converted audio to ComfyDeploy's UGC-MANUAL deployment endpoint
- 4
Poll the API for job completion
- 5
Retrieve and return the output MP4 to the user
MP4 video where the face in the image lip-syncs to the provided audio
Requirements
- ffmpeg installed on the system
Installation
clawhub install pauldelavallaz/ugc-manualRun the CLI command, or paste the prompt into Claude Code, Cursor, Codex or whichever coding agent you use.