video-use edits raw footage into final.mp4 by reading transcripts, not frames

alex_verem · x · 2026-07-23

video-use turns raw footage into a finished edit by converting video into a transcript first, then editing from text instead of frames.

Drop clips into a folder and ask for an edit, and it returns a final.mp4.
It uses speech timing, speakers, and pauses to cut on sentence boundaries rather than mid-word.
The tool can also create animations and graphics overlays, and it remembers project state across sessions.
It runs locally on your own machine; only audio is sent out for transcription via ElevenLabs.
The project is open source and had more than 17,600 GitHub stars at the time of the post.

More from coding & agent