AI Music Generation for the Real World

MuseTree provides a production-ready front-end for OpenAI's MuseNet AI music generator. It targets musicians and producers seeking advanced features and a robust workflow beyond the original research demo, enabling more practical music creation.

How It Works

This project enhances MuseNet by offering a Svelte/TypeScript-based web interface. It implements full tree support for exploring multiple generation paths and tracks, alongside extended controls for genre selection, token length, and temperature. Core features include robust persistence via save/load, WAV/MIDI export/import, and automatic retries for failed requests, addressing limitations of the base MuseNet model for real-world use.

Quick Start & Requirements

• Install: Clone the repository, run npm i, then npm run dev.
• Prerequisites: Node.js environment for npm.
• Online Demo: Available at http://musetree.stevenwaterman.uk.
• Live Development: Streams available on Twitch.tv (Mon-Fri 10:30am-3pm BST).

Highlighted Details

• Advanced Generation: Supports over 600 genres, up to 1000 tokens, and adjustable temperature for diverse outputs.
• Multi-Track & Tree Support: Enables development of multiple concurrent tracks and exploration of numerous generation branches.
• Production Features: Includes WAV/MIDI import/export, full save/load persistence, and automatic request retries.
• Bug Fixes: Addresses issues like persistent notes and generation length limits in the base MuseNet.

Maintenance & Community

MuseTree is under active development, with suggestions welcomed via GitHub issues. The developer live-codes on Twitch.tv, offering real-time interaction.

Licensing & Compatibility

The provided README does not specify a license. Compatibility for commercial use or closed-source linking is undetermined.

Limitations & Caveats

The project is described as being in an "early state" and "active development," indicating potential for ongoing changes and undiscovered issues.