Setup & Installation
Or with the ClawHub CLI, for registry-managed skill folders outside a full OpenClaw workspace:
Version History
- Added explicit required/credential fields for VOICE_AI_API_KEY in the SKILL.md manifest. - Bumped skill version from 0.2.0 to 0.2.1. - Documentation now directly documents VOICE_AI_API_KEY as an explicit requirement/credential.
What This Skill Does
Converts text scripts into voiceovers using Voice.ai TTS, producing numbered audio segments, a stitched master file, YouTube chapter markers, SRT captions, and an HTML review page. Optionally replaces the audio track on an existing video using ffmpeg. Caches segments by content hash so only changed sections re-render on rebuild.
Segment-level caching means editing one section of a long script only re-renders that segment, avoiding full re-renders common in batch TTS tools.
When to use it
- Narrating a YouTube video with chapter timestamps
- Adding captions to a screen recording
- Replacing audio on B-roll footage
- Generating podcast episodes with consistent host voice
- Iterating on course narration without re-rendering unchanged sections
Example Workflow
Here's how your AI assistant might use this skill in practice.
User asks: turn my-script.md into a voiced video using the oliver voice and mux it onto my-recording.mp4
- 1Reads my-script.md and splits it into segments by ## headings
- 2Calls Voice.ai TTS API for each segment, skipping any cached unchanged segments
- 3Stitches segments into master.wav using ffmpeg
- 4Generates chapters.txt, captions.srt, and review.html
- 5Muxes master.wav onto my-recording.mp4, outputting muxed.mp4
out/my-video/muxed.mp4 with AI voiceover, plus master.wav, review.html, chapters.txt, and captions.srt
Requirements
Accounts, API keys, or tools you or your AI assistant may need to set up while using this skill.