PY-XIAOZHI

Lightweight, cross-platform multimodal AI interaction framework. Supports real-time voice, vision recognition, and IoT device control, deployable on desktop and ARM embedded platforms.

AI Voice Interaction

Supports voice input and recognition for intelligent human-computer interaction with natural conversation flow. Built with async architecture for real-time audio processing and low-latency responses.

Vision Multimodal

Supports image recognition and processing for multimodal interaction, understanding image content. Integrates OpenCV camera processing for real-time visual analysis.

MCP Tool Server

Modular tool system based on JSON-RPC 2.0 protocol, with built-in music playback, Bazi fortune-telling, and more. Dynamically extensible tool plugins.

IoT Device Integration

Designed with Thing abstraction pattern, supports smart home device control including lights, volume, temperature sensors. Integrates Home Assistant platform for easy expansion.

High-Performance Audio

Real-time audio transmission based on Opus codec, intelligent resampling, 5ms audio frame interval processing for low-latency, high-quality audio experience.

Cross-Platform

Compatible with Windows 10+, macOS 10.15+, and Linux. Supports GUI and CLI dual modes, adapts to platform-specific audio devices and system interfaces.