Talking Head Recut

You've already got the talking head footage and you want to dress it up with kinetic titles, lower thirds, data callouts, or quote cards that sync to what's being said. This pulls the transcript using local Whisper, lets Claude design timed graphic overlays as individual HTML fragments, then composites everything over your untouched video and renders to MP4 via hyperframes. The workflow is inspect the transcript, draft a storyboard in chat, write each card's HTML directly in the conversation, assemble the final composition, and render. If you just need plain subtitles of the spoken words, use embedded-captions instead. This is for designed graphics that package interview or podcast clips without re-editing the underlying footage.

Install to Claude Code

npx -y skills add heygen-com/hyperframes --skill talking-head-recut --agent claude-code