Skip to content
NewMastershot Studio is open: a video editor with no watermark on anything you export. Try it free

Audio to text

Transcribe a recording, such as an interview, meeting, lecture, or voice memo, to text you can copy or download. Free, with no upload and no limit on length.

Files stay on your device

Drop a recording to turn its speech into text

The speech is recognized on your device, so nothing is uploaded and there's no limit on length.

MP3, M4A, WAV, FLAC, and other audio, or a video with sound.

How to convert audio to text

Open a recording in any common format and Mastershot writes down what's said, in paragraphs, with the time each one starts if you want it. Copy the transcript, or save it as a TXT file or as SRT or VTT subtitles. The speech is recognized in your browser, so the recording stays on your device and there are no minutes to buy.

  1. Step 1

    Choose a recording
    Pick an MP3, M4A, WAV, or other audio file.
  2. Step 2

    Transcribe it
    The speech is written down on your device.
  3. Step 3

    Copy or download
    Save the transcript as TXT, SRT, or VTT.

What to expect

Supported inputs
MP3, M4A, WAV, AAC, OGG, FLAC, AIFF, and other recordings this browser can read, or a video whose speech you want.
What happens
Whisper, a speech recognition model, listens on your device and writes down each word with the time it's said. Paragraphs start where the talk pauses.
Limits
The first time, the browser downloads the 77 MB speech model and keeps it. Transcribing takes about a sixth of the recording's length, so an hour takes about 10 minutes. Clear speech comes out best; music, noise, and people talking over each other make more mistakes. It doesn't say who is speaking.

Private by design

Your video stays on your device

Mastershot processes your file locally in this browser. It does not need to be uploaded to a server, and you do not need an account to use the tool.

  • No upload
  • No signup
  • No watermark

Questions

MP3, M4A, WAV, AAC, OGG, FLAC, and AIFF, and the sound of a video too. If your browser can read the file, it can be transcribed.

Yes, with no sign-up and no limit on length. It runs in your browser, so there's no server time for us to charge you for.

No. The speech model runs in your browser, so the recording and its transcript stay on your device.

99 of them, including English, Spanish, Hindi, Portuguese, French, German, Japanese, and Arabic. It detects the language, or you can choose it.