Video to transcript (text)

Get a written transcript of any video or recording with AI that runs in your browser. Copy-ready paragraphs, or a timed subtitle file. Free, private, no sign-up.

  • AI-powered
Drop a video to add subtitlesor tap to choose a video, or an MP3, M4A or WAV file

How to turn a video into text

  1. Add your file

    Drop a video or audio recording in the box above.

  2. Pick the language

    Keep Detect automatically, or choose the spoken language.

  3. Transcribe

    Press Create subtitles and the text appears as it's heard.

  4. Download

    Correct any word, then press Download text.

A transcript of your video, written on your own device

Sometimes you need the words, not subtitles: quotes from an interview, notes from a lecture, a blog post from a video, or a record of what was agreed in a meeting. This tool listens to a video or audio file and gives you a clean transcript, split into paragraphs where the speaker pauses.

The transcription runs in your browser with Whisper, an open speech model. Your recording isn't uploaded, so it suits confidential calls and anything you'd rather not hand to a transcription service. After the model's one-time download, there's no account and no limit.

The text appears as it's heard. When it's done, read through the lines next to the video, fix anything misheard, and download.

What the transcript looks like

The text file has the spoken words with punctuation and capital letters, and a new paragraph wherever there's a pause of more than two seconds. There are no timestamps in it, so you can paste it straight into a document or email.

If you need timings, download the SRT from the same result. It has the same words with the time each line is spoken, and opens in any text editor.

What it doesn't do

It doesn't label who is speaking, so a conversation comes out as one stream of text. It also writes words as they're said, including the occasional false start, so check a quote against the video before you use it.

Music, laughter and other sounds are left out of the text.

How long it takes

Speed depends on your device and the setting you choose. On a recent laptop, Balanced transcribed a 49-second clip in about 9 seconds when the graphics card could help, and about 15 seconds without it. Phones are several times slower, so use a computer for long recordings.

Keep the tab open while it works. Closing it stops the transcription, because it runs on your device rather than on a server.

What people use it for

Interviews

Pull exact quotes without listening back for the right moment.

Lectures and courses

Turn a recorded class into notes you can search and highlight.

Blogs and newsletters

Start an article from what you already said on camera.

Meetings and calls

Keep a private written record of a recorded conversation.

Tips for the best results

  • Audio files work too: drop in an MP3, M4A or WAV of a podcast or voice note.
  • Choose Accurate for languages other than English, strong accents or poor recordings.
  • If the recording starts with music, choose the language yourself so detection doesn't guess from the music.

Questions

What do I do with the text file?

Open it in any text editor and copy what you need. It's plain text, so it pastes cleanly into Word, Google Docs or an email.

Does it tell me who said what?

No. It writes what's said, but doesn't separate speakers. For an interview, add the names yourself.

Is it really free?

Yes. Every tool is free with no watermark, no sign-up and no daily limit.

More AI tools

All AI tools