🎤

Whisper AI

Real-time speech recognition and translation

Loading whisper-base with WebGPU acceleration for optimal performance. The model (~200 MB) will be cached for faster loading next time.

Everything runs directly in your browser using 🤗 Transformers.js and ONNX Runtime Web with WebGPU support!

Initializing...

Input Language:

Output Language:

Status: Ready

Microphone: 🎤 Off

Original Speech

Speak to see transcription here...

→

Translation

Translation will appear here...

⚙️ Settings

Font Size: 18px

Background Opacity: 90%

Auto-speak translations

Overlay Mode (for streaming)

Google Translate API Key (optional):

📝 Translation History