Hi there! I wanted to share that LiveKit doesn’t come with built-in support for Rasa as a Natural Language Understanding (NLU) integration right out of the box. But don’t worry—you can still connect Rasa with LiveKit by creating a custom plugin using the LiveKit Agents framework. This framework is designed to be flexible and supports adding plugins for AI services like NLU tools. Here’s a simple way to get started with integrating Rasa: 1. Build a plugin module that talks to Rasa’s API for things like intent classification, entity extraction, or managing dialogue. 2. Register your plugin within the LiveKit plugin system. 3. Use Rasa’s HTTP API or SDK to handle user inputs and get back NLU results. 4. Pass those results along to other parts of your system, like an LLM or agent logic, to generate responses. If you’d like some guidance on creating these plugins, the LiveKit Agents repositories and their CONTRIBUTING.md documents are great resources. And please feel free to reach out if you have any questions or need more help—I’m here for you! Sources: agents/README.md | LiveKit integration guides | LiveKit Docs | Plugin System | livekit/livekit_composite | DeepWiki | README.md