Our project aims to develop an application that utilizes various modalities such as text, speech, and facial emotion recognition to analyze and collect user data. The ultimate goal is to create a chatbot and virtual counselor capable of providing personalized recommendations to users to help overcome mental health issues. The application utilizes state-of-the-art models including BERT for text analysis, DeepSpeech for speech recognition, and CNN for facial emotion recognition.
- BERT (for text analysis)
- DeepSpeech (for speech recognition)
- CNN (for facial emotion recognition)
- Hybrid model recommendation system with reinforcement learning
- UI: ReactJS, React Native
- Model: PyTorch
- Backend: Flask
- Database: SQLite
To ensure user privacy, we implement the following techniques:
- Differential privacy
- Federated learning
To enhance security, the following measures are implemented:
- Encryption
- Multi-factor authentication (MFA)
- Clone the repository.
- Set up the backend using Flask and the specified database (SQLite).
- Install the necessary dependencies for the frontend using npm.
- Run the application locally.
Contributions are welcome! Please follow these guidelines:
- Fork the repository.
- Create a new branch for your feature/bug fix.
- Ensure your code follows the project's coding style.
- Submit a pull request.
This project is licensed under the MIT License.