Transcribe video to text free

Upload the video above, wait, and download the transcript. No account, no email, no card. It accepts MP4, MOV, MKV, AVI and WEBM up to 90MB and 20 minutes, and exports plain text plus SRT and VTT subtitle files with nothing held back.

No — you get a transcript plus an SRT or VTT file, not a re-encoded video. Load that file into your editor or upload it alongside the video. Keeping subtitles as a separate file is usually better anyway: viewers can turn them off, search engines can read them, and you can fix a typo without re-exporting. If you do need them burned in, there's an ffmpeg command above that does it in one line.

Usually under a minute once the server is warm. It spins down when idle, so the first run after a quiet period spends about a minute starting up — which means a one-minute clip and a fifteen-minute one often take roughly the same wall time. Upload time is on top of that and depends on your connection.

3GP, AVI, FLV, M4V, MKV, MOV, MP4, MPEG, MPG, WEBM and WMV, up to 90MB and 20 minutes. The file needs an audio track — a silent screen recording has nothing to transcribe.

Yes. SRT and VTT both download without an account, without a watermark, and without a paid tier — export is the feature most free subtitle tools hold back. The limits are 20 minutes per video and 2 per hour.

No — upload the video directly. If you want the audio file itself as well, the Video to Audio converter does that separately.

Extract the audio first with the Video to Audio converter and upload that instead — an audio-only file is a fraction of the size, and the audio is all that gets transcribed anyway.

Yes — choose English output and it translates as it transcribes, in one pass. English is the only translation target, so for any other language you'd translate the finished SRT separately.

Your upload is deleted as soon as processing finishes — not on a timer, and whether the job succeeded or failed. The transcript is available for 1 hour, then removed. Nothing is stored on disk afterwards. There are no accounts, so nothing is linked to you.

Whisper large-v3, running on a GPU. Accuracy depends far more on the recording than on the tool — a lapel mic in a quiet room comes back near-perfect, a phone across a meeting room won't. We don't publish an accuracy percentage, because a single number across every language, accent and recording condition wouldn't mean anything, and nobody publishing one shows their methodology.

SRT for video editors, YouTube, and almost every upload form. VTT for HTML5 video on your own site, via a track element. The only structural difference is the header line and a comma versus a period before the milliseconds.

Usually the wrong format for that target — VTT where SRT was expected, or the reverse. The two look nearly identical in a text editor, and most players fail silently rather than warning you. Try the other export.

Not here — there's no editor. SRT and VTT are plain text, though, so any text editor will do for a quick fix, and every video editor listed above lets you adjust timing and wording after importing.