Real-Time Emotion Detection

Overview

This project aims to detect human emotions in real-time using a webcam feed and a pre-trained deep learning model. The detected emotion is then used to suggest songs based on the emotion detected.

Table of Contents

1. Prerequisites
2. Installation
3. Usage
4. Project Structure
5. Model Training
6. Running the Emotion Detection
7. Spotify Integration
8. Contributions
9. License

Prerequisites

• Python 3.x
• TensorFlow and Keras
• OpenCV
• NumPy
• Spotipy
• Tkinter (for GUI)
• Haar Cascade file for face detection
• Pre-trained emotion detection model (model_optimal.h5)

Installation

1. Clone the repository:

git clone https://github.com/LordAzezl/Real-Time-Emotion-Detection.git

2. Install required Python packages:

pip install numpy tensorflow opencv-python spotipy

3. Download necessary files:

• Download the Haar Cascade file for face detection and place it in the project directory.
• Ensure you have the pre-trained model file (model_optimal.h5) in the project directory.

Usage

Running the Emotion Detection

To run the real-time emotion detection, use the following command:

python emotion_detection.py

Spotify Integration

To use the Spotify integration, you need to have a Spotify developer account. Create a new application on the Spotify Developer Dashboard and obtain the clientID, clientSecret, and redirect_uri. These credentials should be placed in the appropriate section of the script.

Project Structure

Real-Time-Emotion-Detection/
│
├── haarcascade_frontalface_default.xml  # Haar Cascade for face detection
├── model_optimal.h5  # Pre-trained model for emotion detection
├── SongsList/  # Directory containing CSV files with song lists
│   ├── EngHappy.csv
│   ├── EngSad.csv
│   ├── HindiHappy.csv
│   ├── HindiSad.csv
│   └── Allsongs.csv
├── emotion_detection.py  # Main script for emotion detection and Spotify integration
├── model_training.py  # Script for training the emotion detection model
└── README.md  # This README file

Model Training

Data Preparation

• The training and validation datasets should be placed in appropriate directories (train and test respectively).
• The images should be grayscale and of size 48x48 pixels.

Training the Model

The model_training.py script contains the code for training the emotion detection model. To train the model, execute the script:

python model_training.py

This script uses convolutional neural networks (CNNs) to train the model on the provided dataset and saves the trained model as model_optimal.h5.

Model Evaluation

The script evaluates the model on the training and validation datasets and prints the final training and validation accuracy.

Running the Emotion Detection

The emotion_detection.py script captures video from the webcam, detects faces, and predicts the emotion of each detected face in real-time. Based on the detected emotion, it selects a random song from the appropriate CSV file and plays it using the Spotify API.

python emotion_detection.py

Spotify Integration

The script uses the spotipy library to interact with the Spotify API. It authenticates the user and retrieves their profile information. When an emotion is detected, it searches for a song from the corresponding CSV file and opens it in the web browser using Spotify.

Contributions

Contributions are welcome! Please fork the repository and submit a pull request with your improvements.

License

This project is licensed under the MIT License.

---

This README provides an overview of the Real-Time Emotion Detection project, including installation instructions, usage examples, and a description of the code structure.