A customizable AI desktop companion project with character settings, voice conversations, long-term memory capabilities, and sub-1-second response times. Integrates with Live2D models for visual presentation.
One-Minute Overview#
My-Neuro is an open-source project that lets you create your own AI desktop companion with customizable characters and voice conversations featuring sub-1-second response times. The project includes long-term memory, visual recognition, voice cloning, and LLM training capabilities, with integration for various Live2D customizations.
Core Value: Create personalized AI companions with human-like interactive experiences
Quick Start#
Installation Difficulty: Medium - Requires AI/LLM knowledge and Python experience, with multiple components to configure
# Clone the project from GitHub
git clone https://github.com/morettt/my-neuro.git
Is this suitable for me?
- ✅ Need personalized AI assistant: Supports custom character appearance, personality, and voice
- ✅ Gaming interaction: Supports various gaming companion features like "You Draw I Guess", Monopoly, etc.
- ✅ Local deployment: Supports fully local inference without third-party API dependencies
- ❌ Quick deployment: Requires a complex configuration process
- ❌ Low-technical users: Requires AI model knowledge and programming basics
Core Capabilities#
1. Ultra-Low Latency Response - Realistic Conversation Experience#
- Fully local inference with conversation delays under 1 second, approaching real human conversation Actual Value: Smooth and natural conversation experience without common AI delays
2. Long-Term Memory System - Building Deep Relationships#
- Remembers user's key information, personality traits, and preferences Actual Value: AI gets to know the user over time, providing increasingly personalized interactions
3. Multimodal Interaction - Rich Communication Methods#
- Supports voice, text, visual recognition, and various interaction methods Actual Value: Users can choose their most comfortable communication method, and AI can select the best interaction mode based on context
4. Emotional Simulation System - Human-Like Emotional Expression#
- Simulates real human emotional changes with independent emotional states Actual Value: AI responses are no longer cold but carry emotion and personality
5. Gaming Companion Features - Immersive Entertainment Experience#
- Supports multiple gaming companions including "You Draw I Guess", Monopoly, Galgame, Minecraft, etc. Actual Value: AI can serve as a gaming partner, providing new gaming experiences and interactions
Tech Stack & Integration#
Development Language: Python Key Dependencies: GPT-SoVITS (TTS), mindcraft (Minecraft AI), playwright-mcp (web operations), MemOS (memory system) Integration Method: Library/SDK
Maintenance Status#
- Development Activity: Actively developed with a clear feature roadmap from the author
- Recent Updates: Recent new releases planned with more features to be implemented in coming months
- Community Response: Has gained some community support and financial sponsorship
Documentation & Learning Resources#
- Documentation Quality: Basic, with official website guidance
- Official Documentation: 点我进官网 (Official Website)
- Example Code: None, but includes LLM-studio folder with local model inference and fine-tuning guidance