Infinite Radio generates endless, contextually aware music streams, ideal for users seeking dynamic background audio that adapts to their activities. It leverages AI music generation and context detection to create a personalized listening experience.

How It Works

The system combines the Magenta RealTime music model for audio generation with context detection. Context can be derived from the top running processes on a user's machine or from visual analysis via InternVL3, allowing the music genre to shift based on user activity.

Quick Start & Requirements

• Music Model: Run via Docker: docker run --gpus all --network host lauriewired/musicbeats:latest. Access web UI at http://127.0.0.1:8080.
• Prerequisites: Docker with GPU support, NVIDIA GPU with CUDA, NVIDIA Container Toolkit.
• DJ Options:
  • macOS App: Download from releases, run .app, configure IP/port, select Process DJ or LLM DJ. May require screen access permissions.
  • Python Process DJ: python process_dj.py 127.0.0.1 8080
  • Python LLM DJ: Requires LM Studio with InternVL3 or similar model running. python llm_dj.py 127.0.0.1 8080

Highlighted Details

• Real-time music generation powered by Magenta.
• Contextual genre selection via system processes or visual analysis (LLM DJ).
• Cross-platform DJ control (macOS app, Python scripts).
• API available for genre control (POST /genre, GET /current-genre).

Maintenance & Community

The project is actively maintained by LaurieWired. Links to GitHub releases and contributor statistics are provided. Twitter handle @lauriewired is available for updates.

Licensing & Compatibility

Licensed under Apache 2.0, permitting commercial use and modification.

Limitations & Caveats

The LLM DJ requires setting up a separate model server (e.g., via LM Studio), adding complexity to the setup. The macOS application may require specific user permissions for screen analysis.