"Transforming mental wellness through compassionate AI." 🌟
NeuroMentor is an AI-powered mental wellness companion designed for students, professionals, and job-seekers facing stress and emotional challenges.
It blends Generative AI (Gemini 1.5 Pro, Gemini Pro Vision), LangChain, and custom agents to offer private, supportive, and empathetic interactions.
Built during the Google x Kaggle Gen AI Intensive 2025, NeuroMentor showcases real-world GenAI applications like RAG, Few-Shot Agents, Emotional Chatbots, Document Insights, and Voice-enabled Conversations.
| Feature | Description |
|---|---|
| 🧠 Emotion-aware Journaling | Structured output prompts for emotional check-ins |
| 📚 Personal Document Analysis | Upload documents/images for emotional insights |
| 🎙️ Voice Lounge | Speak naturally; AI responds empathetically |
| 🌐 Real-time Web Search | Live, grounded information with web references |
| 🧩 Adaptive Agents | Personalized few-shot prompting and agent-based emotional support |
| 📈 Mood Analytics | Track mood over sessions with visualizations |
| 🔒 Secure & Private | Local session storage; no external sharing |
Mental health is a growing concern in fast-paced environments.
NeuroMentor offers a private, intelligent, always-available companion to guide, motivate, and support mental well-being.
- Frontend: Streamlit (Python)
- Backend: Gemini 1.5 Pro APIs (Text & Vision)
- Agent Framework: LangChain
- Voice Tech: SoundDevice, SpeechRecognition, Pyttsx3
- Vector Search: Chroma VectorDB
- Others: HuggingFace Transformers, Serper.dev for Web Search
🎤 Voice Lounge: Speak freely, NeuroMentor listens and responds supportively.
💬 ChatterBox: Emotion-aware text chat, with positive psychology techniques.
📄 Docs & Images: Upload documents, analyze emotions, and ask questions.
🌐 Search Solutions: Live web search for mental health topics.
📊 History & Insights: Visualize your emotional journey across sessions.
- ✅ Structured Output / JSON Mode
- ✅ Few-Shot Prompting
- ✅ Document Understanding
- ✅ Image Understanding
- ✅ Agents
- ✅ Retrieval Augmented Generation (RAG)
- ✅ Embeddings + Vector Search
- ✅ Audio Understanding (STT and TTS)
- All user data (chats, voice sessions) are saved locally in
data/chat_sessions/by username. - No personal data is sent to external servers except when querying Gemini or Serper.dev APIs.
- No user authentication tokens are hardcoded; sensitive keys are managed securely via
.envfile. - TTS (Text-to-Speech) and STT (Speech-to-Text) processing is done locally where possible.
Unlike a standard chatbot, NeuroMentor offers:
- Natural conversational tone (WhatsApp-like)
- Private journaling and reflection insights
- Continuous emotional memory and history
- Cross-modal support (voice + text + documents)
It brings real-world impact by empowering users to take charge of their emotional health using AI.
1. Clone the Repository
```bash
git clone https://github.com/your-username/neuromentor.git
cd neuromentor
2. Install Dependencies
```bash
pip install -r requirements.txt
3. Configure Environment Variables
Create a .env file:
```bash
GOOGLE_API_KEY=your_google_generativeai_key
SERPER_API_KEY=your_serper_dev_key
✅ Important: Never expose your API keys publicly!
4. Run the Application
```bash
streamlit run app.py
neuromentor/ ├── app.py # Main Streamlit app ├── requirements.txt # Required Python packages ├── .env # Environment variables (keep secret) ├── assets/ # Logos, style.css ├── components/ # Streamlit UI components (home, chat, voice, etc.) ├── utils/ # Utilities (voice_utils, vision_utils, llm, etc.) ├── services/ # User auth and session management ├── agent/ # NeuroMentor agent and tools ├── data/ # User-specific chat history storage └── README.md # This file
Demo: [YouTube Demo Link Here] (https://youtu.be/kuIdqiiyhxM)
Blog: [Medium Blog Link] (https://medium.com/@ayesha24banu/neuromentor-a-genai-powered-mental-wellness-companion-58e01548b2a2)
Demo: NOTE: The audio is working in code and execution, But unable to record audio in this vedio Demo
Neuromentor.gen.Ai.app.1.mp4
Contributions are welcome! Feel free to fork the repo and submit a pull request 🚀.
This project is licensed under the MIT License — free for personal and educational use.
- Google Cloud Gen AI APIs
- Kaggle x Google Gen AI Intensive Course
- HuggingFace Transformers
- Streamlit Community
- Serper.dev API
Through Generative AI and a caring design philosophy, NeuroMentor represents a meaningful step forward in AI-assisted mental wellness.
I’m excited to keep evolving it — making AI not just smart, but emotionally supportive and empowering. 🌟
If you have any questions, suggestions, or collaboration ideas, feel free to reach out:
- Name: Ayesha Banu
- Email: ayesha24banu@gmail.com
- Kaggle: kaggle.com/ayesha990
- LinkedIn: linkedin.com/in/ayesha-banu-cs
- GitHub: github.com/Ayesha24banu
🔔 Feel free to connect for project discussions, hiring, collaborations, or mentorship opportunities!