Tool not found.
Back to Home
Open-source, transcript-based video and audio editor that runs fully offline — delete words in the transcript to cut the matching clip, with no uploads or servers involved.
Rescript is a transcript-based media editor: drop in a video or audio file and it's transcribed locally with per-word timestamps and speaker labels, then editing becomes as simple as deleting words from the transcript — the corresponding clip is cut from the media automatically. All processing happens entirely on-device with no authentication and no file uploads, available as both a browser app and native desktop apps for macOS, Windows, and Linux. Transcription runs via transformers.js using Whisper Base or Whisper Small models (WebGPU with WASM fallback) inside a Web Worker, with speaker diarization handled by pyannote-segmentation-3.0, or users can skip transcription entirely by importing an existing SRT, VTT, or JSON caption file. One-click tools remove filler words ('um', 'uh') and silences of 0.3 seconds or longer, while a full timeline offers waveform display, draggable word timing handles, splitting, and independent cut-edge trimming beyond Whisper's original timestamps. Export options cover video (MP4/WebM, 720p–4K), audio (M4A/MP3/WAV), transcripts (TXT/MD), and subtitles (SRT/VTT/JSON), with frame-accurate re-encoding handled by ffmpeg.wasm. Rescript reports anonymous usage stats and crash reports by default, both disableable in settings. It's built with Next.js, Electron, and TypeScript, and released under the PolyForm Noncommercial License 1.0.0, permitting noncommercial use, modification, and sharing but not commercial resale or redistribution; earlier MIT-licensed releases remain available under MIT for those versions only.
