Back to tools
Private audio transcription
Turn WhatsApp voice notes and audio files into text, right in your browser. Nothing is uploaded to a server. No time limits, no quotas.
Running on: detecting…The model downloads once, then stays cached. Switch models between files, not during a transcription.
Drop your audio files here
.opus · .ogg · .m4a · .mp3 · .wav · .webm, or click to browse
Downloading the model…
Everything runs on your machine
OpenAI's Whisper engine runs through Transformers.js, on WebGPU if your browser supports it, otherwise on WebAssembly. Your files never leave your computer.
The first run takes a little longer
The model downloads once, then loads from your browser cache. Every file after that is transcribed with no download wait.