In-Browser
AI Audio Transcription

Just upload an audio file. Whisper AI runs inference right inside your browser, so your audio is never sent to a server.

Fully private— Audio never leaves your device

Works offline— No network needed after the first model download

Fast WebGPU inference— Several times faster than WASM when a GPU is available

Checking your environment…

Model

Fast and lightweight. Best for clear audio

Language

Speaker diarization

OnBeta

Speakers (0 = auto)

If you know the speaker count, setting it greatly improves accuracy

Drop an audio file, or click to choose

MP3, WAV, M4A, FLAC, OGG, WebM (up to 500 MB)

Your audio is never sent outside the browser

How it works

1
Upload audio
Supports major formats like MP3, WAV, M4A and FLAC. Drag & drop works too.
2
AI transcribes automatically
Whisper AI runs inference inside your browser. The first run downloads the model (~40 MB).
3
Export the results
Download as plain text or SRT subtitles. Speaker diarization is supported.