https://github.com/anis196/sowell_personal_therapist
SoWell uses computer vision libraries (OpenCV and MTCNN) and deep learning (DeepFace) to detect and analyze facial expressions in real-time. The detected emotion influences the AI therapist's tone, built using LangChain with Google's Gemini model. The user interface, styled with Bootstrap and custom CSS, (1k+ lines!)
https://github.com/anis196/sowell_personal_therapist
bootstrap css custom-css dockerfile html javascript mtcnn-face-detection opencv python tensorflow-gpu
Last synced: 17 days ago
JSON representation
SoWell uses computer vision libraries (OpenCV and MTCNN) and deep learning (DeepFace) to detect and analyze facial expressions in real-time. The detected emotion influences the AI therapist's tone, built using LangChain with Google's Gemini model. The user interface, styled with Bootstrap and custom CSS, (1k+ lines!)
- Host: GitHub
- URL: https://github.com/anis196/sowell_personal_therapist
- Owner: Anis196
- License: mit
- Created: 2025-04-11T15:36:38.000Z (18 days ago)
- Default Branch: main
- Last Pushed: 2025-04-11T18:17:27.000Z (17 days ago)
- Last Synced: 2025-04-12T02:15:23.011Z (17 days ago)
- Topics: bootstrap, css, custom-css, dockerfile, html, javascript, mtcnn-face-detection, opencv, python, tensorflow-gpu
- Language: CSS
- Homepage:
- Size: 28.7 MB
- Stars: 0
- Watchers: 1
- Forks: 0
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
- License: LICENSE
Awesome Lists containing this project
README
# SoWell - Your Friendly Anonymous Therapist
Welcome to **SoWell**, a web-based application that serves as your personal, anonymous virtual therapist. This project integrates real-time face and emotion detection using a webcam with a conversational AI powered by Google Generative AI. The interface features a dynamic cityscape animation built with HTML, CSS, and Flask.
## Screenshots
---
## Table of Contents
- [Description](#description)
- [Features](#features)
- [Installation](#installation)
- [Contributing](#contributing)
- [License](#license)
- [Contact](#contact)
---
## Description
SoWell uses computer vision libraries (OpenCV and MTCNN) and deep learning (DeepFace) to detect and analyze facial expressions in real-time. The detected emotion influences the AI therapist's tone, built using LangChain with Google's Gemini model. The user interface, styled with Bootstrap and custom CSS, includes a cityscape background with animated elements and a chat interface.
---
## Features
- 🎭 **Real-Time Emotion Detection**: Analyzes facial expressions via webcam and classifies emotions (happy, sad, angry, fear, disgust, neutral, surprise).
- 💬 **Emotion-Adaptive AI**: Adjusts the therapist's tone based on detected emotions for a personalized experience.
- 🌆 **Dynamic Cityscape**: Animated background with trees, towers, clouds, and street elements.
- 📱 **Responsive Design**: Full-width layout adaptable to different screen sizes.
- 💻 **Chat Interface**: Interactive chat with the AI therapist using Flask and jQuery.
- 🕶️ **Anonymous Support**: Provides a safe, anonymous environment for users.
---
## Installation
1. **Clone the repository:**
```bash
git clone https://github.com/your-username/sowell-therapist.git
Replace your-username and sowell-therapist with your GitHub username and repository name.
2. ***Navigate to the project directory:***
```bash
cd sowell-therapist
3. ***Install dependencies:***
```bash
pip install -r requirements.txt
4. ***Set up your Google API key:***
- Create a .env file in the root directory.
- Add your API key like this:
```bash
GOOGLE_API_KEY=your_api_key_here
#Ensure .env is listed in .gitignore to keep it secure.
4. ***Run the application:***
```bash
python together.py
- Open your browser and go to:
```bash
http://127.0.0.1:5000/
- Allow webcam access for real-time emotion detection.
- Use the chat interface (left side) to interact with SoWell. Type a message and click "Send".
- The right side shows the live video feed with emotion labels and a greeting.
## Contributing
Contributions are welcome! Follow these steps:
### Fork the repository.
1. Create a new branch:
```bash
git checkout -b feature-branch
2. Make your changes and commit:
```bash
git commit -m "Add new feature"
3. Push to your fork:
```bash
git push origin feature-branch
4. Open a pull request describing your changes.
- Please follow the existing code style and add clear comments where necessary.
## License:
This project is licensed under the MIT License.
See the LICENSE file for details.
## Contact:
For questions, suggestions, or to report an issue:
- Email: [email protected]
- GitHub: https://github.com/anis196
- Author: Anis Shaikh
- Last updated: April 11, 2025