Speech to Text
Turn a recording into written text with the Whisper speech model. It detects the language automatically and runs in your browser, so your recording is never uploaded.
Also searched as: transcribe audio · audio to text · mp3 to text · voice to text · transcription
Loading the tool…
How to use it
- 1
Add a recording
Choose or drop an MP3, WAV, M4A or OGG file, or a short video.
- 2
Transcribe
The first time, the speech model downloads once (about 77 MB). After that it works on your device.
- 3
Copy or save
Edit the transcript if needed, then copy it or download it as a .txt file.
Questions about speech to text
Which languages does it understand?
Whisper recognises around 100 languages and detects which one is spoken automatically. Clear English gives the most accurate results.
How long can a recording be?
Up to one hour. Transcription runs on your own device, so long recordings take a while; shorter clips are quick.
Is my recording uploaded?
No. Only the speech model is downloaded; your audio is processed on your device and never leaves it.
Getting an accurate transcript
Transcription accuracy depends mostly on the recording. Clear speech, one person at a time and little background noise give the best results. A phone placed close to the speaker beats a laptop microphone across a room.
The language is detected automatically, but the tool works best with common languages. Names, technical terms and heavy accents may need a quick read-through, and the transcript appears in an editable box so you can fix mistakes before copying it or saving it as a text file.
How it runs and what it needs
The speech model runs on your own device, so your recording never leaves it. The first use downloads the model, about 77 MB, so use Wi-Fi if you are on a limited data plan. Afterwards it works from your browser’s cache. Long recordings take a while, and recordings are limited to one hour; use Audio Trimmer to cut anything longer.
Guides
How to extract audio from a video
Save the soundtrack of a video as an MP3 file. Works with MP4, MOV, WebM and MKV. No software to install.
How to transcribe audio to text for free
Turn a recording into a text transcript with AI that runs in your browser. Free, no sign-up, and your audio stays on your device.