Offline AI-powered transcription using Whisper model - All processing happens in your browser
Enter a YouTube video URL. Audio will be extracted via Invidious.
Privacy-focused YouTube proxy. Try different instances if one fails.
Upload an audio or video file from your device. Supported formats: MP3, MP4, WAV, M4A, WEBM, OGG
Larger models are more accurate but take longer to load and process.
Select language for better accuracy, or use auto-detect.