GhostVideoStudio

Transcribe video & make subtitles.

A transcript or timestamped SRT/VTT subtitles from any video, with OpenAI Whisper running in your browser. The video never leaves your device.

Get a text transcript or subtitle file from any video without uploading it. GhostVideo extracts the audio in your browser with ffmpeg, then runs OpenAI's Whisper speech-recognition model on your device — the video never leaves your computer. Export plain text for notes, or timestamped SubRip (.srt) / WebVTT (.vtt) subtitles ready to drop into an editor or player. The multilingual model handles dozens of languages with optional auto-detection; a faster English-only model is there when you want speed. Perfect for captioning clips, transcribing interviews and webinars, or pulling quotes — including footage too sensitive for a cloud service. Free, no signup, no length cap.

How it works

  1. 1

    Drop your video

    Drag an MP4, MOV, WebM, or other video onto the page or click to browse. It opens in the GhostVideo studio with Transcribe pre-selected.

  2. 2

    Pick format and language

    Choose plain text, SRT, or VTT subtitles, and auto-detect the language or set it. The first run downloads the speech model once, then it's cached.

  3. 3

    Download the transcript

    GhostVideo extracts the audio and transcribes it on your device; the transcript or subtitle file downloads straight from your browser.

Frequently asked questions

  • Is my video uploaded to transcribe it?

    No. The audio is extracted with ffmpeg in your browser and transcribed by an on-device Whisper model — the video is never sent to a server. Only the model download appears in your DevTools Network tab, never your file.

  • Can I generate subtitles (SRT/VTT) for my video?

    Yes. Choose the SRT or VTT format and the transcript includes per-segment timestamps, ready to load as a subtitle track in a video editor or player.

  • What languages does it support?

    The default multilingual model transcribes dozens of languages and can auto-detect the spoken one. For English footage you can pick a smaller, faster English-only model.

  • How long can the video be?

    There's no fixed cap, but transcription runs on your own device, so longer videos take longer and are bounded by your device's memory. Very large files are best trimmed first.

Every GhostVideo tool runs entirely in your browser. Your file is never uploaded — there's no upload endpoint to send it to.