Voice to Instrument uses Spotify's Basic Pitch AI model to detect the notes in your humming or singing. The neural network analyzes pitch in real time and extracts the melody as note events - then the tool re-synthesizes them with your chosen instrument.
Everything runs locally in your browser via ONNX Runtime - your recording never leaves your device. The model is only about 230 KB and downloads instantly. Hum a tune, pick an instrument, and hear it transformed.
Yes - completely free with no limits. The AI runs on your own device.
No. All pitch detection and synthesis happens locally in your browser.
Preview it in the browser and export it as a WAV file to use in your projects.
Piano, guitar, saxophone, and strings - more coming soon.
Hum or sing directly with the record button, or upload any audio file with a single melody line.