Home
Audio Transcriber
Model
Tiny
~40 MB · fastest
Base
~74 MB · balanced
Small
~244 MB · most accurate
Language
Auto-detect
English
Spanish
French
German
Italian
Portuguese
Dutch
Polish
Russian
Japanese
Chinese
Arabic
Hindi
Korean
Options
Timestamps
Models run entirely in your browser via WebAssembly. Downloaded once from HuggingFace, then cached — no data leaves your machine.
Drop audio here or click to browse
MP3, WAV, M4A, OGG, FLAC, WebM — up to 500 MB
Transcribe
Record
Transcript
Copy