This project provides a cross-platform AI assistant client built with Flutter, supporting real-time voice and text interactions with multiple AI models like Xiaozhi, Dify, and OpenAI. It targets users who want a unified interface for various AI services on iOS, Android, Web, and desktop platforms.

How It Works

The client leverages Flutter for its cross-platform capabilities, enabling a single codebase for multiple operating systems. It communicates with backend AI services primarily through WebSockets for real-time voice conversations and also supports text and image-based interactions. The architecture allows for flexible configuration of multiple AI service endpoints and authentication methods.

Quick Start & Requirements

• Install dependencies: flutter pub get
• Run application: flutter run
• System Requirements: Flutter ^3.7.0, Dart ^3.7.0, iOS 12.0+, Android API 21+ (5.0+), Modern Web Browsers.
• Official Demo Video: https://github.com/TOM88812/xiaozhi-android-client/blob/main/demo.gif

Highlighted Details

• Supports multiple AI models including Xiaozhi (WebSocket), Dify, and OpenAI (with image and streaming output).
• Features real-time voice calls with manual interruption and push-to-talk functionality.
• Includes automatic reconnection mechanisms and mixed voice/text chat history.
• Offers configuration for multiple AI service addresses and API keys.

Maintenance & Community

The project is actively developed, with recent updates including new versions and feature implementations. The developer encourages community contributions, particularly for iOS echo cancellation.

Licensing & Compatibility

The repository does not explicitly state a license. Compatibility for commercial use or closed-source linking is not specified.

Limitations & Caveats

iOS echo cancellation is not yet implemented. Dark/light theme adaptation is planned but not yet available. Some advanced features like local ASR, wake word, and TTS are still in the development pipeline.