xiaozhi-esp32-server and py-xiaozhi
Project B (py-xiaozhi) appears to be a *complement* to Project A (xinnan-tech/xiaozhi-esp32-server), as B provides a software-only "Xiaozhi AI" experience, while A provides the backend service for hardware-based ESP32 device control, suggesting that B could potentially interact with or leverage the services offered by A for a more complete or integrated smart home experience.
About xiaozhi-esp32-server
xinnan-tech/xiaozhi-esp32-server
本项目为xiaozhi-esp32提供后端服务，帮助您快速搭建ESP32设备控制服务器。Backend service for xiaozhi-esp32, helps you quickly build an ESP32 device control server.
Based on the README, here's a technical summary: Implements intelligent device control through MQTT+UDP and WebSocket protocols using a Python/Java/Vue stack, supporting streaming ASR/TTS pipelines for sub-3-second response latency. Features voice biometrics, knowledge base integration, and MCP endpoints for multi-agent management, with flexible deployment options—from minimal file-based configurations to full database-backed architectures supporting concurrent user sessions and OTA updates.
About py-xiaozhi
huangjunsen0406/py-xiaozhi
A Python-based Xiaozhi AI for users who want the full Xiaozhi experience without owning specialized hardware.
Implements multimodal AI voice interaction with offline wake-word detection via Sherpa-ONNX and MCP tool ecosystem (schedule management, music search, map services, IoT device control). Built on asyncio event-driven architecture with WebSocket/MQTT protocol support, WebRTC audio processing, and PyQt5 GUI alongside CLI mode for cross-platform deployment.
Related comparisons
Scores updated daily from GitHub, PyPI, and npm data. How scores work