EDUMCP

EDUMCP, short for Educational Model Context Protocol, is an innovative open platform protocol designed to break down the silo effect currently seen in AI applications within education. Built on the robust Model Context Protocol (MCP), EDUMCP enables seamless interconnection and interoperability among diverse AI models, educational applications, smart hardware, and even teaching content.

An AI tutoring system can easily share and understand data and context with a story generation model, an interactive game engine, or a virtual experiment platform. By establishing such a standardized communication bridge, EDUMCP brings unprecedented possibilities to the field of education, enabling various AI tools to work together to provide users with richer, more personalized, and more inspiring learning content and services.

EDUMCP One-Sentence Video Creation Framework

You only need to provide a one-sentence topic, and EDUMCP can generate a roughly 30-60 second animated video for children. It will support translation of English: dialogue content modification and image changes. like：Create an video that EDUMCP robot that makes the ROYA start to enjoy learning.

Overview

AICHILDEDU is an innovative educational platform designed specifically for children, leveraging the power of AI to create engaging, personalized, and educational content. The platform uses a microservice architecture to deliver a variety of AI-powered educational experiences, allowing for flexible scaling and feature expansion.

Features

Personalized Story Generation: Create custom educational stories tailored to specific age groups, themes, and educational focuses
Educational Quiz Creation: Generate engaging quizzes and questions that reinforce learning objectives
Multimedia Integration: Combine text, images, voice, and video into comprehensive educational materials
Age-Appropriate Content: Content customization based on age groups and developmental stages
Multi-Language Support: Support for content generation in multiple languages
User Management: Comprehensive user system with parent, teacher, and admin roles
Parental Controls: Robust parental control features to ensure child-appropriate content
Asynchronous Processing: Non-blocking task execution for resource-intensive AI operations

System Architecture

EDUMCP follows a microservice architecture with the following key components:

                           ┌─────────────────┐
                           │   API Gateway   │
                           └───────┬─────────┘
                                   │
┌────────┬────────┬────────────────┼────────────────┬────────────────┐
│        │        │                │                │                │
▼        ▼        ▼                ▼                ▼                ▼
┌──────────┐ ┌─────────┐ ┌──────────────┐ ┌──────────────┐ ┌──────────────┐
│   User   │ │ Content │ │ Learning     │ │Recommendation│ │  Analytics   │
│ Service  │ │ Service │ │ Service      │ │   Service    │ │   Service    │
└──────────┘ └─────────┘ └──────────────┘ └──────────────┘ └──────────────┘
                                 ┌──────────────────────────┐
                                 │       AI Services        │
                                 │                          │
                                 │ ┌─────────┐ ┌─────────┐  │
                                 │ │  Text   │ │ Image   │  │
                                 │ │Generator│ │Generator│  │
                                 │ └─────────┘ └─────────┘  │
                                 │ ┌─────────┐ ┌─────────┐  │
                                 │ │  Voice  │ │  Video  │  │
                                 │ │Generator│ │Generator│  │
                                 │ └─────────┘ └─────────┘  │
                                 └──────────────────────────┘

Technologies Used

Backend: Python, FastAPI, Uvicorn
Databases: PostgreSQL, MongoDB, Redis
AI & ML: LangChain, EDUMCP GPT models, TensorFlow, PyTorch, Transformers
Storage: MinIO (Object Storage)
Search: Elasticsearch
Containerization: Docker, Docker Compose
Authentication: JWT-based authentication

Getting Started

Prerequisites

Docker and Docker Compose
Python 3.10 or higher (for local development)

Installation

Clone the repository:

git clone https://github.com/CHILDEDUAI/aichildedu.git
cd aichildedu

Create an .env file in the root directory with the following variables:

# Database
POSTGRES_USER=postgres
POSTGRES_PASSWORD=postgres
MONGODB_URI=mongodb://mongodb:27017/
MONGODB_DB=EDUMCP

# Authentication
SECRET_KEY=your_secret_key_change_in_production

Start the services using Docker Compose:
```
docker-compose up -d
```
The API Gateway will be available at http://localhost:8000

Local Development Setup

Create a virtual environment:

python -m venv venv
source venv/bin/activate  # On Windows, use `venv\Scripts\activate`

Install dependencies:
```
pip install -r requirements.txt
```

Run the desired service:

uvicorn aichildedu.user_service.main:app --reload --port 8001

Usage Examples

Generating a Story

import requests

api_url = "http://localhost:8000/api/v1/ai/text/story"

story_request = {
    "title": "The Curious Robot",
    "theme": "Technology and Friendship",
    "age_group": "6-8",
    "characters": [
        {
            "name": "Robo",
            "description": "A curious and friendly robot who wants to learn about the world"
        },
        {
            "name": "Mia",
            "description": "A smart girl who loves technology and building things"
        }
    ],
    "educational_focus": "Introduction to robotics and programming concepts",
    "length": "medium",
    "language": "en"
}

response = requests.post(api_url, json=story_request)
task = response.json()

print(f"Story generation task created: {task['task_id']}")
print(f"Check status at: {task['status_check_url']}")

Checking Task Status

import requests

task_id = "task_12345"
status_url = f"http://localhost:8000/api/v1/ai/text/tasks/{task_id}"

response = requests.get(status_url)
status = response.json()

print(f"Task status: {status['status']}")
print(f"Progress: {status['progress']}%")

Retrieving Generated Content

import requests

task_id = "task_12345"
result_url = f"http://localhost:8000/api/v1/ai/text/tasks/{task_id}/result"

response = requests.get(result_url)
story = response.json()

print(f"Story Title: {story['title']}")
print(f"Summary: {story['summary']}")
print("\nContent:")
print(story['content'])

API Documentation

The API documentation is available at:

API Gateway: http://localhost:8000/docs
User Service: http://localhost:8001/docs
Content Service: http://localhost:8002/docs
Learning Service: http://localhost:8003/docs
AI Text Generator: http://localhost:8010/docs
AI Image Generator: http://localhost:8011/docs
AI Voice Generator: http://localhost:8012/docs
AI Video Generator: http://localhost:8013/docs

Service Architecture

User Service

Handles user management, authentication, and authorization. Manages user profiles, child accounts, and parental controls.

Content Service

Manages educational content creation, storage, retrieval, and organization with the following features:

Content Types Management:
- Stories: Interactive educational narratives with character development, themes, and moral lessons
- Quizzes: Customizable question sets with answers, difficulty levels, and scoring mechanisms
- Lessons: Structured learning materials with educational objectives and prerequisites
Content Organization:
- Categories: Hierarchical organization of content with parent-child relationships
- Tags: Flexible content labeling for improved discoverability and filtering
- Collections: User-created or curated sets of content (custom, curriculum, featured)
Content Filtering and Search:
- Age-appropriate filtering based on min/max age ranges
- Difficulty level filtering (beginner, intermediate, advanced)
- Content rating filtering (G, PG, educational)
- Multi-language support
- Full-text search capabilities
Content Metadata:
- Educational value tracking
- Subjects and themes classification
- Reading time estimation
- Word count metrics
User Interaction:
- Content reactions (like, favorite, helpful)
- Content sharing and permissions
- User-specific content collections
Multimedia Asset Management:
- Storage and retrieval of images, audio, and video assets
- Asset association with educational content

Learning Service

Tracks learning progress, personalized learning paths, and educational achievements with the following features:

Learning Path Management:
- Curriculum Design: Structured learning paths based on age groups and subjects
- Prerequisites: Dependency management for learning materials
- Progress Tracking: Real-time monitoring of learning milestones
- Adaptive Learning: Dynamic adjustment of content difficulty based on performance
Assessment and Evaluation:
- Progress Assessment: Regular evaluation of learning outcomes
- Skill Level Analysis: Tracking of subject-specific competencies
- Performance Metrics: Detailed analytics of learning activities
- Achievement System: Gamified rewards and badges for accomplishments
Personalization Features:
- Learning Style Detection: Identification of individual learning preferences
- Content Recommendations: AI-powered suggestions for next learning steps
- Pace Adjustment: Flexible learning speed based on individual progress
- Interest-Based Learning: Content tailoring based on demonstrated interests
Progress Reporting:
- Detailed Analytics: Comprehensive learning progress visualization
- Parent/Teacher Dashboard: Monitoring tools for guardians and educators
- Progress Reports: Regular automated assessment summaries
- Learning Insights: AI-generated recommendations for improvement
Collaborative Learning:
- Peer Learning: Facilitated group learning activities
- Social Interaction: Safe, moderated peer communication
- Group Projects: Collaborative educational tasks
- Shared Achievements: Community learning milestones
Learning Support:
- AI Tutoring: Intelligent assistance for difficult concepts
- Homework Help: Guided problem-solving support
- Review Sessions: Automated review of challenging materials
- Learning Resources: Additional materials for deeper understanding

AI Services

Text Generator

Creates educational stories and quizzes with the following features:

Personalized story generation based on age, theme, and educational focus
Quiz generation with customizable difficulty levels
Asynchronous task processing for better resource management
Template-based content customization

Image Generator

Creates illustrations to accompany educational content:

Story illustration generation
Character visualization
Educational diagrams and charts

Voice Generator

Provides audio narration for educational content:

Story narration with different character voices
Multi-language support
Age-appropriate voice adaptation

Video Generator

Creates educational videos and animations:

Story animations
Educational concept visualizations
Interactive learning content

Development and Contribution

Development Workflow

Fork the repository
Create a feature branch
Implement your changes
Write tests for new features
Submit a pull request

Code Style

This project follows PEP 8 style guidelines. Please ensure your code adheres to these standards before submitting PRs.

License

This project is licensed under the Apache License 2.0 - see the LICENSE file for details.

edumcp