The words of a video, as text.
Drop a video or a recording. You get a clean transcript in paragraphs, with or without times, to copy into a script, a caption or a blog post.
Runs in your browser. Nothing is uploaded, no signup.
Watch an example
14 seconds · a made-up example
The studio's Clips mode reads the transcript of your own long recordings and picks the moments that stand alone as shorts.Join the beta
How it works
Short and checkable: every limit links to the platform's own page, read on 28 Oct 2026.
Heard on your computer
Whisper, OpenAI's open speech model, runs in your browser through Transformers.js. The sound is cut into parts of at most 30 seconds at quiet moments, and each part is recognised with the time of every word.
Laid out like a person writes
A new paragraph starts where the speaker pauses for more than 1.2 seconds, or at the end of a sentence once a paragraph is long. Spaces before punctuation are mended (French keeps its space before ; : ! ?), and the first letter gets its capital.
Tidy on request
Filler sounds (um, uh, euh, hmm) can be left out with one tick, and the times shown or hidden. Every time is a button that plays that paragraph.
What it doesn't do
- It can mishear names, brands and numbers: read it through before you publish it.
- Two people talking at once come out as one stream of words; it doesn't say who speaks.
- Up to 30 minutes per file.
- Nothing is sent or saved. The file stays in this tab; only the model is downloaded, once.
Questions
How do I transcribe a video for free?
Drop it here and press Write the transcript. The speech is recognised in your browser by Whisper (OpenAI's open model), laid out in paragraphs where the speaker pauses, and you copy it or save it as a TXT file. No upload, no account, no limit.
Which languages does it understand?
Whisper recognises dozens of languages; the list offers the common ones and Detect. The Accurate model (77 MB) makes fewer mistakes than the Fast one (41 MB), especially on names and numbers.
Can I turn the transcript into subtitles?
Yes: the auto-subtitles tool uses the same engine and cuts the words into timed subtitles (SRT, VTT) or burns them into the video.
Sources
- 1Hugging Face · onnx-community — Whisper tiny and base with word timestamps (ONNX builds of OpenAI's Whisper)Apache-2.0; downloaded by your browser · read 28 Oct 2026
- 2Hugging Face — Transformers.js 3.8.1 — runs the model in your browserApache-2.0 · read 28 Oct 2026
More free tools
All 25 tools- Make the videoShort video analyzerWhy does this short work?
- Make the videoScript timerHow long will this script run?
- Make the videoAuto subtitlesNo subtitles yet? Make them here.
- Make the videoBurn subtitlesSubtitles that stay on TikTok and Reels?
- Make the videoSubtitle checkerCan people read your subtitles in time?
- Make the videoSilence removerCut the pauses out of a talking video?
From a long recording to the shorts in it.
Vertiscale's Clips mode reads the transcript of your own long recordings, picks the moments that stand alone and makes each one a vertical short.