AI Music Generator
Generate music from text prompts using MusicGen. Powered by Transformers.js - runs entirely in your browser.
ℹ️ Using Quantized Models: This tool uses Q8 quantized models via Transformers.js for better browser compatibility. Models are automatically downloaded and cached on first use.
ℹ️ Technical Details (click to expand)
- Model: MusicGen-small (Xenova/musicgen-small via Transformers.js)
- First Use: Models will be downloaded and cached in your browser (IndexedDB)
- Privacy: All processing happens locally in your browser
- Format: Models are in ONNX format, loaded via Transformers.js
- Output: Generates audio in WAV format at 32kHz sample rate
Models not loaded. Click "Generate Music" to download and load models (first time only).
Generated audio will appear here...
FAQ
Is my data sent to a server?
No. All processing happens locally in your browser. The models are downloaded once and cached locally.
How long does generation take?
Generation time depends on the duration and your device. Typically 10-30 seconds for 10 seconds of audio on a modern device.
What audio format is generated?
The tool generates audio in WAV format at 32kHz sample rate.
Can I use this commercially?
Please check the MusicGen model license on Hugging Face. The model is typically available under Apache 2.0 license.