About this project

AI Jam Sessions is an MCP server that provides LLMs with sensory-like capabilities to learn and experience music. It moves beyond simple MIDI libraries by offering a comprehensive teaching environment for piano, guitar, and vocals. Key features include: - **Sound Engines**: Six distinct audio engines including additive synthesis for piano and guitar, physical modeling for vocal tracts, and sample-based playback (including a sampled Concert Grand). - **Visual & Analytical Tools**: A piano roll that renders music as SVG for AI verification, interactive guitar tablature, and a browser-based "Cockpit" for composition, tuning, and A/B auditions. - **Song Library**: 120 songs across 12 genres (Blues, Classical, Jazz, etc.), each with deep musical annotations covering harmonic analysis and performance tips. - **Learning Loop**: A persistent practice journal that tracks progress across sessions, allowing the AI to reflect on patterns and set goals. - **jam-actions-v0 Dataset**: A public dataset of 115 multi-turn MCP tool-use traces grounded in classical piano MIDI, designed to train LLMs in symbolic music analysis. The system provides 49 MCP tools categorized into Learn, Play, Practice, and Sing, enabling AI to list songs, analyze measures, play through speakers, and score performances.