Add audio
Drop an existing file or capture a fresh microphone recording in the browser.
Turn MP3 files and voice recordings into text.
Drag in an audio file or record directly from your microphone. Your audio stays on your device while a browser-based Whisper model creates the transcript.
🔒 No audio uploads🎙 Direct recording⬇ TXT, SRT and VTT downloads
Ready to record
00:00
The first transcription downloads the selected AI model to the browser cache. Larger models can improve accuracy but need more memory and download time.
Transcript
Your transcript will appear here
Select or record audio, choose your settings, then start transcription.
Preparing transcription…
Loading the speech-recognition model.
0%
Process locally
The selected Whisper model runs on your device using browser-based machine learning.
Edit and export
Review the transcript and save it as plain text, subtitles or structured JSON.
Privacy and browser requirements
Your audio is processed in browser memory and is not intentionally uploaded by this website. Model files are downloaded from Hugging Face/CDN infrastructure and cached by your browser. Microphone recording requires permission and usually HTTPS. Transcription speed and accuracy depend on audio quality, language, model size and your device.