ableton -extended

Ableton MCP Extended provides programmatic control over Ableton Live using natural language via an AI assistant. The server exposes a rich set of tools for session, transport, track, clip, device, automation, browser, and voice/audio integration. With this MCP server, you can start and stop playback, query session information (tempo, time signature, track count), create and fire scenes, manage tracks and devices, modify MIDI clips and notes, load instruments and effects by URI, and import audio directly into the session. It also includes features for text-to-speech narration or spoken elements through ElevenLabs integration, voice cloning for character voices in your tracks, and extensible tools for custom controllers with ultra-low latency networking. You interact with these capabilities by sending structured JSON commands from your AI assistant to the MCP server, which translates them into Ableton Live actions via the remote script.

To use it, configure the MCP server in your AI assistant’s MCP config with the command path to the server (for example, python MCP_Server/server.py). Then issue natural language requests like: create a new MIDI piano track, load a piano instrument, or add a drum clip and quantize it. The assistant will map your intent to the appropriate MCP tool call, and the server will execute the corresponding Ableton Live action through the remote script embedded in Ableton Live.

Prerequisites:

• Ableton Live 11+ and a compatible OS (Windows or macOS)
• Python 3.10 or higher
• Access to the Ableton Live Remote Script setup as described in the repository

Installation steps:

1. Clone the repository and install the package in editable mode:

git clone https://github.com/uisato/ableton-mcp-extended.git
cd ableton-mcp-extended
pip install -e .

2. Install the Ableton Remote Script as documented:

• Locate Ableton Remote Scripts folder (Windows or Mac paths in the README)
• Create the AbletonMCP folder and copy the remote script into it

3. Ensure Ableton Live is configured to allow the MCP server to communicate via the remote script (Preferences > Link, Tempo & MIDI, set Control Surface to AbletonMCP and configure I/O as described)
4. Run the MCP server from the project using Python:

python MCP_Server/server.py

5. Add the MCP server to your AI assistant’s configuration (example provided in the repository).

Tips and common considerations:

• The ElevenLabs integration is optional and can be wired with another MCP server to generate and import audio directly into Ableton Live. If you enable it, ensure the appropriate API keys and environment variables are configured as described in the ElevenLabs integration notes.
• The UDP/hybrid server features offer ultra-low latency real-time parameter control; refer to the Advanced Features section for setup steps.
• Some automation point features are noted as not working perfectly yet; plan to validate and adjust automation data as needed.
• When loading instruments or effects by URI, verify the URI compatibility with Live’s browser and ensure related samples are accessible on disk or network paths.
• For best results, run the MCP server on the same machine as Ableton Live or ensure the network path latency is minimized if remote connections are used.