Overview

Mitra is an AI-powered mental wellness assistant that provides users with emotional support and personalized self-care plans based on sentiment analysis of chat interactions and facial emotion detection. The backend is built using Flask, incorporating MongoDB for user and chat storage, VADER Sentiment Analysis for mood tracking, Facial Emotion Recognition for music generation, and FPDF for generating self-care reports.

Features

• User Authentication (Registration, Login, Google Authentication, Password Recovery)
• Chatbot API (LLM-based mental health chatbot with sentiment-aware responses and conversation memory)
• Sentiment Analysis (Tracks user emotions and adjusts chatbot behavior accordingly)
• Facial Emotion Detection (Analyzes user facial expressions to detect emotions)
• Music Generation (Creates personalized music based on detected emotions using MusicGEN)
• Self-Care Plan Generation (Creates PDF reports based on sentiment trends)
• Secure API Endpoints (JWT authentication, request validation, and security best practices)

Project Structure

Team-45_Mitra/
│── app/
│   │── data/
│   │   │── haarcascade.xml     # Facial detection model
│   │   │── model.h5            # Emotion classification model
│   │   │── spotify_data.csv    # Music preference data
│   │── routes/
│   │   │── __init__.py         # Route initialization
│   │   │── __pycache__/        # Python cache files
│   │   │── auth_routes.py      # Handles user authentication (login, register, password reset)
│   │   │── chatbot_routes.py   # Chatbot API for generating responses and sentiment analysis
│   │   │── test_routes.py      # Routes for testing purposes
│   │   │── user_routes.py      # Handles user-related actions like profile retrieval
│   │── static/generated_pdfs/  # Storage for generated self-care PDFs
│   │── utils/
│   │   │── __init__.py         # Utility initialization
│   │   │── config.py           # Configuration settings (DB connections, API keys, etc.)
│   │   │── mail.py             # Email handling for password recovery & self-care plan delivery
│   │   │── models.py           # Defines database models for users and chats
│   │   │── security.py         # Security utilities (password hashing, JWT verification)
│── frontend/
│   │── node_modules/           # Frontend dependencies
│   │── public/                 # Static assets
│   │── src/
│   │   │── assets/             # Images, icons, and other static resources
│   │   │── components/         # Reusable UI components
│   │   │── pages/              # Page-level components
│   │   │── After_Login.css     # After login page styles
│   │   │── App.css             # Global styles
│   │   │── App.jsx             # Main application component
│   │   │── App.test.js         # Application tests
│   │   │── breathing.css       # Breathing exercise component styles
│   │   │── Chatbot.css         # Chatbot component styles
│   │   │── Contact_Us.css      # Contact page styles
│   │   │── FAQs.css            # FAQ page styles
│   │   │── index.css           # Index page styles
│   │   │── index.jsx           # Application entry point
│   │   │── logo.svg            # Application logo
│   │   │── meditation.css      # Meditation component styles
│   │   │── MusicAnimation.css  # Music animation component styles
│   │   │── Profile.css         # User profile styles
│   │   │── reportWebVitals.js  # Web vitals reporting
│── README.md                   # Project documentation
│── requirements.txt            # Dependencies need to install before running

Installation & Setup

1. Clone the Repository

git clone https://github.com/your-repo/Team-45_Mitra.git
cd Team-45_Mitra

2. Create a Virtual Environment & Install Dependencies

python -m venv venv
source venv/bin/activate  # On Windows use `venv\Scripts\activate`
pip install -r requirements.txt

3. Configure Environment Variables

Create a .env file in the root directory and add the following:

MONGO_URL="YOUR_MONGO_URL"
JWT_SECRET_KEY="supersecretkey31"
MAIL_SERVER="smtp.gmail.com"
MAIL_PORT=587
MAIL_USE_TLS="True"
MAIL_USERNAME="YOUR_BUISNESS_GMAIL"
MAIL_PASSWORD="AUTHENTICATION PASSWORD"
MONGO_DB_NAME="mydatabase"
CLIENT_ID='YOUR_CLIENT_ID'
CLIENT_SECRET='YOUR_CLIENT_ID'
TOGETHER_API_KEY="YOUR_TOGETHER_API_KEY"

4. Run the Flask Application

python run.py

The server will start at http://127.0.0.1:5000

API Endpoints

Authentication Routes

Method	Endpoint	Description
POST	/api/v1/register	User registration
POST	/api/v1/login	User login
POST	/api/v1/forgot-password	Sends password reset email
POST	/api/v1/reset-password/<token>	Resets user password
GET	/api/v1/login/google	Authorization by google

Chatbot Routes

Method	Endpoint	Description
POST	/api/v1/api/chat	Generates chatbot response with conversation memory & sentiment analysis
POST	/api/v1/api/generate_selfcare_pdf	Generates and emails self-care plan PDF
POST	/api/v1/api/detect_emotion	Analyzes facial expression to detect emotion
POST	/api/v1/api/generate_music	Creates personalized music based on detected emotion

User Routes

Method	Endpoint	Description
GET	/api/v1/get-username	Fetches user username
POST	/api/user/update-profile	Updates user profile
GET	/api/v1/profile	Fetches user details

Key Components

• init.py: Initializes Flask, JWT, Mail, OAuth, and database connections.
• config.py: Manages app configurations via environment variables.
• models.py: Defines MongoDB collections (users, chats).
• Chatbot Memory: Uses LangChain's ConversationBufferMemory to store and reference past conversations.
• Facial Emotion Detection: Uses a pre-trained model to detect emotions from user's facial expressions.
• MusicGEN Integration: Generates emotion-based music using Meta's MusicGEN pre-trained model.

MITRA Frontend

Overview

The MITRA Frontend is a React-based user interface designed for the MITRA AI-powered mental wellness assistant. It serves as the primary platform where users can interact with the chatbot, access therapy recommendations, and explore self-care resources including breathing exercises, meditation guides, and emotion-based music.

Tech Stack

• Frontend Framework: React.js
• Styling: Tailwind CSS
• State Management: Context API (or Redux if used later)
• Routing: React Router
• Authentication: JWT-based authentication (with cookies)
• Backend Communication: REST API (Flask-based backend)

Project Structure

frontend/
│── public/              # Static assets (favicon, logo, index.html)
│── src/
│   ├── assets/         # Images, icons, and other static resources
│   ├── components/     # Reusable UI components (Navbar, Footer, Buttons)
│   ├── pages/          # Page-level components (Home, Login, Profile, Chatbot)
│   ├── App.jsx         # Main application entry point
│   ├── App.css         # Global styles
│   ├── breathing.css   # Breathing exercise component styles
│   ├── meditation.css  # Meditation component styles
│   ├── MusicAnimation.css # Music player with animation styles
│── package.json        # Dependencies and scripts
│── tailwind.config.js  # Tailwind CSS configuration
│── .env                # Environment variables (API URLs, secrets)

Features

• User Authentication (Login, Register, Forgot Password)
• Chatbot Integration (AI-powered mental wellness assistant with conversation memory)
• Facial Emotion Recognition (Camera-based emotion detection)
• Music Therapy (AI-generated music based on detected emotions)
• Self-care Tools (Breathing exercises, meditation, and therapy plans)
• Responsive UI with Tailwind CSS
• Dark Mode Support (if implemented)
• Secure API Communication (JWT token authentication via cookies)

Installation & Setup

1. Clone the repository:

   git clone https://github.com/smitngandhi/Team-45_Mitra.git
   cd frontend

2. Install dependencies:

   npm install

3. Start the development server:

   npm start

Environment Variables

Create a .env file in the root directory and define:

REACT_APP_API_BASE_URL=http://127.0.0.1:5000/api/v1
REACT_APP_GOOGLE_CLIENT_ID=your_google_client_id

API Endpoints Used

• User Authentication:
  • POST /api/v1/login
  • POST /api/v1/register
• Chatbot:
  • POST /api/v1/api/chat
• Emotion Detection:
  • POST /api/v1/api/detect_emotion
• Music Generation:
  • POST /api/v1/api/generate_music

Contributing

1. Create a new branch: git checkout -b feature-name
2. Make changes and commit: git commit -m 'Added new feature'
3. Push to GitHub: git push origin feature-name
4. Open a pull request

Maintainers

• Smit Gandhi (@smitngandhi)
• Dhruvil (@Dhruvil-Joshi)
• Prachi (@PrachiDesai2506)