Musicgen Small

🚀 MusicGen Small with Transformers.js

This project makes the facebook/musicgen-small model compatible with Transformers.js using ONNX weights, enabling music generation in the browser.

🚀 Quick Start

This repository contains the facebook/musicgen-small model with ONNX weights, making it compatible with Transformers.js.

📦 Installation

If you haven't already, you can install the Transformers.js JavaScript library from NPM using:

npm install @huggingface/transformers

💻 Usage Examples

Basic Usage

Generate music with Xenova/musicgen-small.

import { AutoTokenizer, MusicgenForConditionalGeneration } from '@huggingface/transformers';

// Load tokenizer and model
const tokenizer = await AutoTokenizer.from_pretrained('Xenova/musicgen-small');
const model = await MusicgenForConditionalGeneration.from_pretrained('Xenova/musicgen-small', {
  dtype: {
    text_encoder: 'q8',
    decoder_model_merged: 'q8',
    encodec_decode: 'fp32',
  },
});

// Prepare text input
const prompt = 'a light and cheerly EDM track, with syncopated drums, aery pads, and strong emotions bpm: 130';
const inputs = tokenizer(prompt);

// Generate audio
const audio_values = await model.generate({
  ...inputs,
  max_new_tokens: 500,
  do_sample: true,
  guidance_scale: 3,
});

// (Optional) Write the output to a WAV file
const audio = new RawAudio(audio_values.data, model.config.audio_encoder.sampling_rate);
audio.save('musicgen.wav');

You can listen to a sample audio here:

We also released an online demo, which you can try yourself: MusicGen Web Demo

You can watch a demo video here:

📄 License

This project is licensed under the cc-by-nc-4.0 license.

📚 Documentation

⚠️ Important Note

Having a separate repo for ONNX weights is intended to be a temporary solution until WebML gains more traction. If you would like to make your models web-ready, we recommend converting to ONNX using 🤗 Optimum and structuring your repo like this one (with ONNX weights located in a subfolder named onnx).