Linxi

🎙️ Free tool · no sign-up

Private transcription that never leaves your device

Turn a recording into text with an open speech model that runs inside your browser. Your audio is never uploaded, stored or seen by anyone.

  • Audio never leaves your device
  • No account, no limit per day
  • Text and subtitle downloads

How to use it

Speech to text in three steps

  1. Choose a recording

    Drop in an audio or video file, or record from your microphone.

  2. Let it listen

    The first time, your browser fetches the speech model once. Then it transcribes, 30 seconds at a time.

  3. Take the text

    Copy it, or download it as a text file or as subtitles (.srt) for a video.

Why it is private

How in-browser transcription works

Most transcription sites upload your audio to their servers. This one sends your browser the model instead.

0Bytes of audio uploaded

Your recording is decoded and transcribed in your browser. It never travels over the internet.

AIWhisper, an open speech model

The same open model used across the industry, in a compact version built to run in a browser.

41 MBFetched once

The fast model is about 41 MB and the better one about 77 MB. Your browser keeps it for next time.

30 minPer recording

Up to 30 minutes at a time. Longer recordings can be split into parts first.

Questions

Transcription questions

Is this transcription really private?

Yes. The audio is decoded and transcribed by a speech model running in your own browser. It is never uploaded, and nothing is stored on our servers.

How accurate is it?

Clear speech in a quiet room comes out well. Choose “Better” for harder recordings. Names and jargon may need a quick edit.

Which languages does it support?

English by default. Choose “Detect (any language)” to transcribe dozens of other languages with the multilingual model.

Can I make subtitles for a video?

Yes. Drop in the video file, then download the .srt file, which most video players and editors accept.

Why is the first run slower?

The browser fetches the speech model the first time. After that it is kept on your device, so the next recording starts right away.

Does it work on a phone?

It works best on a computer. Newer phones can run it, but long recordings may be slow.

Start free

Private by design

Linxi keeps what is yours on your side: captions on your own device, end-to-end encrypted messages, and no selling of data.