Skip to content
Projects

TOM88812/xiaozhi-android-client

View on GitHub

TOM88812/xiaozhi-android-client has 1,659 GitHub stars and 423 forks. GitHot measured +2 stars in 24 hours. Last checked 2026-11-07 03:57 UTC. Last push 2026-09-18.

How these counts are measured

一个基于小智、xiaozhi-server的Android、IOS语音对话应用,支持实时语音交互和文字对话。现在是flutter版本,打通IOS、Android端。请同志们动动小手,点点小星星,予以鼓励。

Measured star growth

in 24h
+2
in 7 days
Collecting history
in 15 days
Collecting history
in 1 month
Collecting history

Net star changes over overlapping periods, not daily totals. Missing measurements are left blank. Stars measure attention, not software quality.

README_EN.md

Original

Xiaozhi AI Assistant Android / iOS Client

Release License: Apache-2.0 Stars Download

The new version has been released, please experience it! Echo cancellation for Flutter iOS and Android has been implemented. PRs are welcome. If you find this project useful, feel free to donate. Every donation is my motivation to move forward. Dify supports sending image interactions. Multiple Xiaozhi agents can be added to the chat list.

Xiaozhi AI Assistant developed based on the Flutter framework, supporting multi-platform (iOS, Android, Web, Windows, macOS, Linux) deployment, providing real-time voice interaction and text conversation functions.

New Version
New Version iOS, Android Client (Web and PC versions can be built manually)
Click to watch demo video

V3 Commercial Version Features (Deeply Adapted to Self-developed Server) 💼

Feature Module Status Description
Adaptive Theme ✅ Dark/Light theme adaptation/Follows device theme
AI Service Provider ✅ Supports OpenAI services, using LLMs on mobile
Thinking Mode ✅ Supports OpenAI Thinking Mode
HTML Code Preview ✅ Models write simple HTML code for preview, AI programming on mobile
MCP_Client ✅ Supports MCP capability calls, interface data can be DIYed
OpenAI Interface Online Search ✅ Supports OpenAI interface service online search
Video Playback ✅ Supports playing videos returned by models
OpenAI Speed Test ✅ OpenAI interface response speed test, service speed at a glance
Live2D ✅ Multi-model switching, supports importing your favorite model characters
IoT ✅ Supports calling phone functions, navigation, music listening, etc.
Innovative Mood Mode ✅ Supports real-time interruption mode
MQTT-UDP ✅ Supports MQTT protocol service, long connection
WS ✅ Supports WS protocol service
Real-time Voice Interruption ✅ Interrupt at will while Xiaozhi is speaking, say what you want unstoppable
Multiple Xiaozhi Services ✅ Add multiple Xiaozhi services, easily implement multiple assistants per person
Hardware Integration ✅ Interconnect with hardware, separate memories
Deep Adaptation to Server ✅ Adapted to commercial version server
User Information ✅ Display membership expiration time, conversation count, bound device count, voiceprint count, quota usage, recent active devices
Device Management ✅ Supports mobile viewing of all devices for the current logged-in user, adding new devices
Role Management ✅ Supports mobile management of your current role, creating new roles
Voiceprint Management ✅ Supports mobile voiceprint recording, making your AI understand you better
Conversation History ✅ Supports displaying recent conversation records
Memory Management ✅ Supports displaying memories
Reserved Pages ✅ Reserved pages for bookkeeping, todo, diary, etc. UI, can integrate Xiaozhi backend capabilities extension, collaborate with Xiaozhi to build an assistant that understands you

Server Commercial Version Features 💼

Feature Module Status Description
First Sentence Response ✅ Wake word response time <1s, extreme speed response experience
Average Response Speed ✅ Average conversation response time <2s (Public CDN network) (Local intranet limit within 800ms), smooth conversation experience
MQTT Protocol ✅ Supports MQTT communication protocol, long connection, server active wake-up
Voice Cloning ✅ Supports Volcengine voice cloning, realizing personalized voice customization
Voiceprint Recognition ✅ Supports voiceprint recognition function, realizing personalized voice assistant
Bi-directional Streaming Interaction ✅ Supports Volcano streaming playback, real-time voice input and reply output
User Client ✅ Friendly user client operation interface, native card-style device management page
MCP Access Point ✅ Role-based MCP tool access point, extension function access (Sync Xiage Service)
MCP Hub Service ✅ SSE/HTTP MCP Hub support for more third-party service integration
ESP32 Device Theme Customization ✅ Supports online configuration of ESP32 device themes and sticker packs
Function Call ✅ Tool calling, improving user experience
Long-term Memory ✅ Extract key information records based on user dialogue, intelligent memory management
Monitoring Panel ✅ Monitor Token, conversation duration, device activity, etc. data in day, week, month dimensions
OTA Firmware Upgrade ✅ Firmware upload, automatic upgrade, remote device management
Chat Data Visualization ✅ Chat frequency statistical charts and other data visualization functions, monitoring conversation data trends
User Membership System ✅ Supports setting Token count based on membership level, supports monthly/yearly subscription, online payment (Alipay, WeChat, PayPal)

Contact Information

  • email

[email protected]

  • wechat

Tang_xs-xk

For customized client development support, please contact WeChat

Server Graphical Deployment Tool

🌟 Support

Your every star⭐ or donation💖 is our motivation to keep moving forward🛸.

Donate

ko-fi

Star History

Explore related projects