🧠 VoiceWithin

A mental health companion web app for students — journaling, AI chat, risk assessment, and mood tracking, all in one place.

VoiceWithin is a Flask-based web application designed to support student mental health. It combines a mood journal, an AI-powered chatbot, a mental health risk assessment survey, and push notifications into a single, empathetic platform.

---

✨ Features

🤖 AI Chatbot

• Powered by Groq's LLaMA 3.3 70B model
• Responds in short, warm, Gen-Z-friendly tones — like a supportive friend at 2AM
• Crisis detection: instantly surfaces the AASRA helpline (🇮🇳 9820466627) if distress signals are detected
• Short-term memory: retains the last 10 messages per chat for context
• Auto-names chats based on detected emotions (e.g. "Stressed Talk", "Happy Chat")
• NLTK VADER sentiment analysis runs on every user message

📓 Mood Journal

• Write daily entries with a title, content, and mood tag
• Attach images (stored in MongoDB GridFS)
• View, edit, and delete past entries
• Mood statistics with daily and monthly trend summaries

📊 Mental Health Risk Assessment

• Multi-question survey covering depression, anxiety, panic attacks, suicidal ideation, self-harm, bullying, substance use, and more
• ML model (scikit-learn) predicts risk level: Low / Medium / High
• Tailored suggestions based on prediction
• History tracked per user for the improvement dashboard

📈 Improvement Dashboard

• Combines QA risk history and mood journal data
• Shows progress over time with visual charts

🔔 Push Notifications

• Web Push via VAPID keys
• Motivational messages sent at 10 AM and 7 PM daily using APScheduler

---

🗂️ Project Structure

VoiceWithin/
├── hie.py                    # Main Flask app (routes, chatbot, ML, notifications)
├── fixed_chat_route.py       # Isolated chat route (dev iteration artifact)
├── requirements.txt          # Python dependencies
├── Procfile                  # Render deployment config
├── suicide_model(3).pkl      # Trained ML model (scikit-learn)
├── label_encoders.pkl(3)     # Label encoders for ML features
├── .gitignore
└── templates/
    ├── intro.html            # Landing page
    ├── index.html            # Signup page
    ├── login.html            # Login page
    ├── main_menu.html        # Dashboard
    ├── survey.html           # Mental health QA form
    ├── QA.html               # QA interface
    ├── result.html           # Prediction result
    ├── journal.html          # New journal entry
    ├── hist.html             # Journal history
    ├── chatbot.html          # AI chat interface
    └── improve.html          # Progress dashboard

---

🛠️ Tech Stack

Layer	Technology
Backend	Python, Flask
Database	MongoDB Atlas (PyMongo, GridFS)
AI / LLM	Groq API — LLaMA 3.3 70B
ML Model	scikit-learn (pickle)
Sentiment	NLTK VADER
Push Notifications	pywebpush + APScheduler
Auth	Flask sessions
Deployment	Render (free tier)
Frontend	HTML, CSS, JavaScript

---

🚀 Getting Started

Prerequisites

• Python 3.9+
• MongoDB Atlas account
• Groq API key (get one here)
• VAPID keys for Web Push (generate with py-vapid or web-push-codelab)

1. Clone the repo

git clone https://github.com/Sayedani-29/VoiceWithin.git
cd VoiceWithin

2. Install dependencies

pip install -r requirements.txt

You may also need to download the NLTK lexicon:

import nltk
nltk.download("vader_lexicon")

3. Set up environment variables

Create a .env file in the root directory:

GROQ_API_KEY=your_groq_api_key_here
MONGO_URI=your_mongodb_connection_string_here
SECRET_KEY=your_flask_secret_key_here

⚠️ Never commit your .env file. It's already included in .gitignore.

4. Run the app

python hie.py

Visit http://localhost:5000 in your browser.

---

☁️ Deployment (Render)

This project is configured for deployment on Render.

The Procfile tells Render how to start the app:

web: python hie.py

Steps:

1. Push your code to GitHub.
2. Create a new Web Service on Render and connect your repository.
3. Add the following environment variables in Render's dashboard:
   • GROQ_API_KEY
   • MONGO_URI
   • SECRET_KEY
4. Deploy.

💡 The app includes a keep-alive thread that pings itself every 14 minutes to prevent Render's free tier from sleeping.

---

🔑 Environment Variables

Variable	Description
GROQ_API_KEY	API key for Groq LLM access
MONGO_URI	MongoDB Atlas connection string
SECRET_KEY	Flask session secret key

---

🧩 API Routes

Method	Route	Description
GET	/	Intro / landing page
GET/POST	/index	Signup
GET/POST	/login	Login
GET	/logout	Logout
GET	/main_menu	Dashboard (auth required)
GET	/survey	Mental health survey
POST	/predict	Run ML prediction
GET	/result	Show prediction result
GET	/journal	New journal entry form
POST	/add	Save journal entry
GET	/hist	Journal history
POST	/edit/<id>	Edit a journal entry
POST	/delete/<id>	Delete a journal entry
GET	/mood_stats	Mood statistics (JSON)
GET	/improve	Improvement dashboard
GET	/chatbot	Chatbot UI
POST	/chat	Send a message to the bot
POST	/save_subscription	Register push notification subscription
GET	/health	Health check endpoint

---

🤝 Contributing

Contributions are welcome! Here's how to get involved:

1. Fork this repository
2. Create a branch for your feature or fix:

   git checkout -b feature/your-feature-name

3. Make your changes and commit with a clear message:

   git commit -m "feat: add your feature description"

4. Push to your fork:

   git push origin feature/your-feature-name

5. Open a Pull Request against the master branch

Ideas for contribution

• Add password hashing (bcrypt)
• Add email verification on signup
• Build a more detailed progress dashboard
• Add multilingual support
• Improve mobile responsiveness

---

🐛 Known Issues & Security Notes

These are known limitations to be aware of before deploying to production:

• Passwords are stored in plaintext — use bcrypt or werkzeug.security to hash passwords before storing.
• VAPID keys are hardcoded in hie.py — move them to environment variables.
• Sessions are not encrypted beyond Flask's default — set a strong, random SECRET_KEY.
• fixed_chat_route.py is a development artifact and is not used in production.

---

💛 Acknowledgements

• Groq — blazing fast LLM inference
• NLTK — sentiment analysis
• MongoDB Atlas — cloud database
• AASRA — crisis helpline integrated for Indian users (24/7: 9820466627)
• Everyone who contributed to making mental health tools more accessible 💛

---

📄 License

© 2024 Sayedani-29. All rights reserved. This project and its contents may not be reproduced, distributed, or used without explicit permission from the author.

"Your story matters. You matter." 🌻