Audio and video to text, free and private by default

Transcribe interviews, lectures, podcasts and videos for free. By default the AI runs inside your browser, so your recording stays with you. Need higher accuracy, for example in Urdu? Switch on Cloud mode: only the sound is sent to be transcribed, and nothing is stored.

  • Private by default: your file stays on your device
  • No sign-up, no minute limits, no watermark
  • Around 100 languages, with translation to English

🔒 Private mode: your file stays on this device

How it works

  1. Choose your file

    Pick an audio or video file from your computer or phone. Your browser opens it right on your device.

  2. The AI listens

    By default it runs on your device: the first time, your browser downloads the speech model and keeps it. For hard languages or on phones, optional Cloud mode uses a larger AI instead.

  3. Edit and download

    Fix any word in the transcript, then copy it or download it as TXT, SRT or VTT.

Why it's free and private

Most transcription sites send your file to their servers, and every minute costs them money. That is why they limit free use, add watermarks or ask you to subscribe.

This site works differently. By default the speech recognition model, OpenAI's open-source Whisper, runs inside your own browser. Your file is never uploaded and we never see it, so there is nothing to limit. If you switch on optional Cloud mode for higher accuracy, only the sound is sent to our server, transcribed and sent straight back. It is never stored, and Cloud mode has a generous free daily allowance.

Questions

Is it really free?

Yes. There is no sign-up and no watermark. On-device transcription has no limits because each file costs us nothing. Optional Cloud mode is free too, with a shared daily allowance.

Is my file uploaded anywhere?

Not by default. Your browser reads the file and the AI processes it on your device; only the AI model is downloaded, once. If you switch on Cloud mode, the sound (never the video picture) is sent to our server to be transcribed. It is not stored and not used to train AI. Leave Cloud mode off and nothing leaves your device.

Which languages does it understand?

Around 100, including English, Spanish, Hindi, Urdu, Arabic, Portuguese and Indonesian. Choosing the spoken language usually gives better results than automatic detection.

How long can my file be?

On your device there is no fixed limit: longer files take longer, and speed depends on your device. Cloud mode handles files up to 30 minutes. Keep this tab open until it finishes.

Which browser works best?

Chrome or Edge on a computer, because they can use your graphics card. Other browsers and phones work too, but more slowly.

Can I send a video to it from another app?

Yes, on Android. Open the site in Chrome and tap Install the app (or the menu, then Add to Home screen). After that, in your Gallery, WhatsApp, Files or any app, tap Share and choose Quietscribe: the video opens here ready to transcribe, without uploading anything. On iPhone, tap Choose a file: it opens your Photos and Files straight away.