Audio to text transcription in your browser
Free audio transcription online, powered by Whisper. Turn voice notes, interviews and meetings into text right in your browser. No upload, no sign-up, export .txt.
100% private — files never leave your deviceYour browser can’t run this tool. Please try the latest Chrome, Edge or Safari on a desktop.
Loading the AI model (first time only)…
The AI model downloads once, then runs entirely on your device — this can take a few seconds on the first run. It’s cached afterwards, so next time is instant.
How to use
- Add your audio. Drop in a voice memo, interview or meeting recording, or pick a file from your device.
- Choose a model size. Pick a smaller Whisper model for speed or a larger one for better accuracy; it downloads once on first use.
- Run the transcription. Click transcribe and let the model process the audio locally while the text builds up on screen.
- Review and export. Read through the result, fix any names or jargon, then export a clean .txt file.
Features
- Nothing leaves your device. Whisper runs locally in the browser, so your recordings are never uploaded to a server.
- Works offline after setup. Once the model is cached on first use, you can transcribe again with no internet connection.
- Free with no limits or sign-up. Transcribe as many clips as you like without accounts, watermarks or per-minute charges.
- Choose accuracy vs speed. Smaller models transcribe fast on modest hardware; larger ones catch more detail in tricky audio.
- Clean .txt export. Download the finished transcript as a plain text file you can edit, search or paste anywhere.
When to use it
- Turn a recorded interview into a searchable text transcript for an article or report.
- Convert voice memos and dictated notes into written text you can clean up later.
- Get a rough text record of a meeting or lecture you recorded on your phone.
- Draft subtitles or captions from a video's exported audio track.
- Transcribe a podcast episode so you can repurpose quotes and show notes.
Is it private?
Your audio is transcribed locally in your browser with Whisper and is never uploaded to any server.
Frequently asked questions
Is this audio transcription tool free?
Yes, it is completely free. There is no sign-up, no subscription and no per-minute charge, and you can transcribe as many files as you want.
Are my recordings uploaded anywhere?
No. The Whisper model runs entirely in your browser, so your audio is processed on your own device and never sent to a server.
What audio formats and lengths are supported?
Common formats like MP3, WAV, M4A and similar work well. There is no hard length limit, but very long recordings use more memory and can be slow or fail on low-memory phones, so splitting long files helps.
Why does the first transcription take a moment to start?
The first run downloads the Whisper model once (a few hundred MB depending on the size you choose). After that it is cached, so later transcriptions start faster and even work offline.
How accurate is the transcription?
Clear speech in a quiet room transcribes very well. Heavy accents, overlapping voices, background noise and uncommon names are harder, so expect to proofread and fix a few words, especially with smaller models.
Does it work offline?
Yes, once the model has been downloaded and cached on first use, you can transcribe without an internet connection.
What languages can it transcribe?
Whisper supports many languages and can usually detect the spoken language automatically, though accuracy is highest for widely spoken languages and clear audio.
Last updated