py-xiaozhi

py-xiaozhi

Open-source AI assistant ecosystem supporting MCP integration, multimodal workflows, IoT, and cross-platform voice interaction.

MCPDevOpen sourceAI助手多模态IoT跨平台
Type
MCP
Transport
http
Open source
Yes
GitHub Stars
★ 3.4k
Source
mcp-github

Overview

py-xiaozhi is a lightweight, cross-platform multimodal AI interaction framework built on Python's asynchronous architecture. It supports real-time voice streaming, vision-language tasks, and IoT device control. Can be deployed on Windows, macOS, Linux desktops, and ARM embedded platforms (e.g., Raspberry Pi), seamlessly connecting large language models with physical hardware. The framework offers multiple UI modes including PySide6 + QML GUI, CLI, and GPIO interface, suitable for desktop, headless server, and embedded environments.

Capabilities

  • Real-time voice AI with low latency
  • Multimodal vision with image understanding
  • MCP tool ecosystem with modular JSON-RPC 2.0
  • Cross-platform deployment supporting multiple operating systems
  • Multiple UI modes adapting to different environments
  • Offline wake-word recognition

Use cases

Smart home controlRobotics controlMultimodal AI application developmentCross-platform AI assistant

Setup

Requires: Python 3.10 - 3.12音频设备网络连接
pip install py-xiaozhi

This information was compiled by AI from public sources and may contain inaccuracies — please refer to the source.

FAQ

Supports Windows 10+, macOS 10.15+, Linux (x86_64 & ARM).

Supports Windows 10+, macOS 10.15+, Linux (x86_64 & ARM).

Requires microphone and speaker devices; recommends at least 4GB RAM and modern CPU.

Requires microphone and speaker devices; recommends at least 4GB RAM and modern CPU.

Related skills