xiaozhi-esp32

xiaozhi-esp32

An AI chatbot based on the MCP protocol, supporting multi-device control.

MCPOfficeOpen source
Type
MCP
Transport
stdio
Open source
Yes
GitHub Stars
★ 27.5k
Source
mcp-github

Overview

XiaoZhi AI Chatbot is a voice interaction interface based on the MCP protocol, leveraging the AI capabilities of large models such as Qwen/DeepSeek to enable multi-device control. It supports Wi-Fi and 4G communication, offline voice wake-up, multiple communication protocols, audio codecs, a streaming ASR+LLM+TTS architecture, speaker recognition, OLED/LCD display, battery management, and multilingual support. Compatible with ESP32-C3, ESP32-S3, and ESP32-P4 chip platforms. Device-side MCP handles device control (speaker, LED, servo motor, GPIO, etc.), while cloud-side MCP extends large model capabilities (smart home control, PC desktop operations, knowledge search, email, etc.).

Capabilities

  • Wi-Fi/4G communication
  • Offline voice wake-up
  • Websocket/MQTT+UDP protocol
  • OPUS audio codec
  • Streaming ASR+LLM+TTS architecture
  • Speaker recognition

Use cases

Smart home controlPC desktop operationKnowledge searchEmail management

Setup

Requires: API KeyNode环境
We recommend beginners use pre-built firmware without requiring a development environment setup for flashing.

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

FAQ

How to upgrade to version 2?

Manually flash the firmware to upgrade to v2.

Which hardware platforms are supported?

Supports ESP32-C3, ESP32-S3, ESP32-P4, and other chip platforms.

Related skills