Fork of DataAnts-AI/CutScript. The original project is the source of essentially everything here — the Electron shell, React frontend, WhisperX transcription, FFmpeg export pipeline, and AI integrations are all DataAnts-AI's work. This fork adds features on top and contributes improvements back upstream where the maintainers want them. See Docs/attribution.md for details.
An open-source, local-first text-based audio and video editor powered by AI. Edit your video by editing the transcript — delete a word and it's cut from the video.
- Text-based editing — select and delete transcript words to cut video
- Bulk copy/paste editing — copy the transcript, edit in any text editor, paste back to apply all cuts at once
- Word-level transcription — WhisperX with speaker diarization
- AI assistance — filler word removal and clip suggestions via Ollama, OpenAI, or Claude
- Studio Sound — background noise removal with DeepFilterNet
- Export — stream-copy (fast) or re-encode up to 4K, with caption burn-in
- Full undo/redo, keyboard shortcuts, waveform timeline, project save/load
npm install
cd frontend && npm install && cd ..
cd backend && uv sync && cd ..
npm run devSee Docs/getting-started.md for detailed setup, prerequisites, and troubleshooting.
| Topic | Description |
|---|---|
| Getting Started | Install, dev setup, prerequisites |
| Architecture | Tech stack and project structure |
| Features | Full feature reference |
| Keyboard Shortcuts | All keyboard shortcuts |
| API Reference | Backend REST endpoints |
| Attribution | Fork history and what changed |
MIT — see LICENSE for details.