ugc-manual_skill

Overview

This skill generates a lip-synced MP4 video by combining a single image (with a visible face) and a user-provided audio recording. It preserves the exact timing and natural nuances of the supplied voice or audio track. The output is a ready-to-use video where the subject in the image moves lips to match the audio.

How this skill works

The skill accepts an image (JPG/PNG) and an audio file (MP3/WAV/OGG or similar). It auto-converts input audio to the required WAV PCM format, sends media to a lipsync pipeline, and renders an MP4 with synchronized mouth movements. Processing time typically ranges from 2–5 minutes depending on audio length and complexity.

When to use it

• You have a user-recorded voice file and need exact timing preserved in the video.
• You want to turn a single portrait or character image into a speaking video.
• You have pre-generated audio (TTS or human recording) and need matching lip-sync.
• You need to reuse or archive user-submitted audio combined with an image.
• You want to avoid AI voice generation and keep the original speaker's voice.

Best practices

• Use a clear image with a visible frontal or 3/4 face for best results.
• Provide high-quality audio with minimal background noise for cleaner lip motion.
• Trim silence at the start/end of the audio to avoid unnecessary mouth movement.
• Install ffmpeg so the skill can reliably convert audio formats before processing.
• Test with short clips first to verify visual alignment and audio fidelity.

Example use cases

• Create personalized greeting videos using a submitted voice note and a profile photo.
• Convert podcast excerpts or voice messages into short social clips with a character image.
• Localize an influencer's image by swapping in user-recorded translations or dubs.
• Produce marketing snippets where a brand avatar speaks exact recorded lines.
• Archive user-generated audio paired with a static image for social sharing.