Is xiaozhi-esp32-server-java free to use? What license does it have?

xiaozhi-esp32-server-java is open source under the MIT license. Check the license terms before using it in commercial products.

Is xiaozhi-esp32-server-java still maintained?

Actively maintained. The last commit to xiaozhi-esp32-server-java was on Sep 27, 2026.

What can I use xiaozhi-esp32-server-java for?

Running a self-hosted voice assistant backend for ESP32 devices in a home or office Managing a fleet of Xiaozhi-compatible devices with OTA upgrades and monitoring Building a voice assistant that controls smart-home devices through MCP or function calls Prototyping an AI voice product on ESP32 hardware with swappable speech and LLM providers

How do I install xiaozhi-esp32-server-java?

Create a folder and enter it: mkdir xiaozhi && cd xiaozhi Download the docker-compose.yml file: curl -O https://raw.githubusercontent.com/joey-zhou/xiaozhi-esp32-server-java/main/docker-compose.yml Start the containers: docker compose up -d Once the containers are healthy, open http://localhost:8084 and log in as admin / 123456, then configure an LLM API key

Where can I get xiaozhi-esp32-server-java?

The source code and installation instructions are on GitHub: https://github.com/joey-zhou/xiaozhi-esp32-server-java

joey-zhou/xiaozhi-esp32-server-java

Java server and admin platform for Xiaozhi ESP32 voice devices, covering device monitoring, voice and persona management, and conversation logs.

View on GitHub

Stars: 1,381
Forks: 505
Open issues: 2
Last commit: 23 days ago

What is it?

Xiaozhi ESP32 Server Java is a Java backend for ESP32-based voice hardware compatible with the Xiaozhi ESP32 firmware. It ships with a web admin console for managing users, devices, roles and OTA upgrades, and a separate dialogue service that handles WebSocket and MQTT audio streams and the AI conversation pipeline.

The system is split into two processes, xiaozhi-server for administration and xiaozhi-dialogue for real-time voice, which share MySQL and Redis and can be scaled independently. Speech recognition and synthesis can use local models such as sherpa-onnx and Vosk or cloud providers, and language models can come from OpenAI, Zhipu, Ollama, Dify, Coze and others. MCP tool support, function calling, a RAG knowledge base and voiceprint recognition extend the assistant's abilities.

Deployment is available through Docker Compose or from source with JDK 21, Maven, Node 22, MySQL 8 and Redis 7. Models are downloaded by a provided script on first run.

Key features

01Separate admin and dialogue processes that share MySQL and Redis and scale independently
02Local and cloud speech recognition and synthesis, including voice cloning and streaming interaction
03Integration with OpenAI, Zhipu, Ollama, Dify, Coze and other LLM providers
04MCP tool protocol support and function calling for smart-home control
05RAG knowledge base for PDF, Office and image documents
06Long-term memory and voiceprint recognition for multiple household members

Use cases

Running a self-hosted voice assistant backend for ESP32 devices in a home or office
Managing a fleet of Xiaozhi-compatible devices with OTA upgrades and monitoring
Building a voice assistant that controls smart-home devices through MCP or function calls
Prototyping an AI voice product on ESP32 hardware with swappable speech and LLM providers

Quick start

1
Create a folder and enter it
mkdir xiaozhi && cd xiaozhi
2
Download the docker-compose.yml file
curl -O https://raw.githubusercontent.com/joey-zhou/xiaozhi-esp32-server-java/main/docker-compose.yml
3
Start the containers
docker compose up -d
4
Once the containers are healthy, open http://localhost:8084 and log in as admin / 123456, then configure an LLM API key

Things to know

You must supply your own LLM API key before the system can hold conversations; no keys are included.
Some features are not open source and are only available in the commercial version.

Topics

#esp32 #java #spring-boot #voice-assistant #iot #llm #speech-recognition #text-to-speech #mcp #mcp-client

FAQ

What is xiaozhi-esp32-server-java?: Java server and admin platform for Xiaozhi ESP32 voice devices, covering device monitoring, voice and persona management, and conversation logs. Xiaozhi ESP32 Server Java is a Java backend for ESP32-based voice hardware compatible with the Xiaozhi ESP32 firmware. It ships with a web admin console for managing users, devices, roles and OTA upgrades, and a separate dialogue service that handles WebSocket and MQTT audio streams and the AI conversation pipeline.
Is xiaozhi-esp32-server-java free to use? What license does it have?: xiaozhi-esp32-server-java is open source under the MIT license. Check the license terms before using it in commercial products.
Is xiaozhi-esp32-server-java still maintained?: Actively maintained. The last commit to xiaozhi-esp32-server-java was on Sep 27, 2026.
What can I use xiaozhi-esp32-server-java for?: Running a self-hosted voice assistant backend for ESP32 devices in a home or office Managing a fleet of Xiaozhi-compatible devices with OTA upgrades and monitoring Building a voice assistant that controls smart-home devices through MCP or function calls Prototyping an AI voice product on ESP32 hardware with swappable speech and LLM providers
How do I install xiaozhi-esp32-server-java?: Create a folder and enter it: mkdir xiaozhi && cd xiaozhi Download the docker-compose.yml file: curl -O https://raw.githubusercontent.com/joey-zhou/xiaozhi-esp32-server-java/main/docker-compose.yml Start the containers: docker compose up -d Once the containers are healthy, open http://localhost:8084 and log in as admin / 123456, then configure an LLM API key
Where can I get xiaozhi-esp32-server-java?: The source code and installation instructions are on GitHub: https://github.com/joey-zhou/xiaozhi-esp32-server-java

Summary generated by AI from the project's README. GitHub stats refresh daily.