Xiaozhi Esp32
XiaoZhi ESP32 is an open-source voice interaction entry point powered by large language models such as Qwen and DeepSeek. It utilizes the Model Context Protocol (MCP) to enable multi-terminal control, allowing the AI to interact with speakers, LEDs, servos, GPIO pins, and extend capabilities for smart home control, PC desktop operations, knowledge search, and email management. The system supports Wi-Fi and ML307 Cat.1 4G connectivity, featuring offline voice wake-up via ESP-SR, streaming ASR, LLM integration, and TTS. Audio is handled using the OPUS codec. The device includes speaker recognition through 3D Speaker technology, supports OLED and LCD displays with emojis, and features battery management. It is compatible with ESP32-C3, ESP32-S3, and ESP32-P4 chip platforms and offers multi-language support for Chinese, English, and Japanese. Users can customize wake words, fonts, emojis, and backgrounds via an online web-based editor. The firmware supports both Websocket and MQTT plus UDP communication protocols