Back to Apps

Xiaozhi AI Backend
by hackers365
High-performance streaming AI backend for ESP32 devices with ASR, LLM, TTS, and OpenClaw integration.
0 stars
Works in:claude
Exposes:ToolsResources
What it does
Xiaozhi AI Backend is a Go-based, full-streaming AI server designed for IoT and smart voice scenarios, specifically targeting ESP32 devices. It orchestrates the entire voice interaction pipeline from speech-to-text to text-to-speech with minimal latency.
Tools
- MCP Audio Server: Manages audio resource pagination and streaming for music playback and volume control.
- OpenClaw Integration: Provides dedicated agent endpoints with keyword-based routing for seamless agent switching.
- Voice Identification: Supports speaker identification and dynamic TTS switching based on user identity.
- Knowledge Base: Integrates RAG capabilities via Dify, RAGFlow, and WeKnora.
Installation
Add the following to your claude_desktop_config.json:
{
"mcpServers": {
"xiaozhi-server": {
"command": "./xiaozhi_server",
"args": ["-c", "config/config.yaml"]
}
}
}
Supported hosts
- claude
Quick install
go build -o xiaozhi_server ./cmd/server/ && ./xiaozhi_server -c config/config.yamlInformation
- Pricing
- free
- Published
- 7/13/2026
- stars
- 0
Categories
Choose your AI client and follow the steps below.
Claude Desktop
JSON for claude_desktop_config.json





