Emotionfy

Emotion-Driven Playlist Recommendation System 🎵

INTRODUCTION

Emotionfy is a cutting-edge system that merges the power of emotion detection with personalized music recommendations. Using deep learning for facial emotion recognition and Spotify’s API, this application recommends playlists tailored to the user’s emotional state. Upload an image, and Emotionfy will do the rest — analyzing your emotion and curating a playlist to match your mood.

TEAM MEMBERS

Rey Reyes: CNN Researcher, Backend Developer
Lucas Yao: Backend Developer, Frontend Support
Fei Lin: Frontend Developer, Backend Support

DATASETS USED

1. Emotion Detection FER Dataset
This dataset contains 35,887 grayscale 48x48 pixel images labeled with the following emotions: - Angry - Disgust - Fear - Happy - Neutral - Sad - Surprise

2. Face Expression Recognition Dataset
A complementary dataset designed for facial emotion recognition tasks.

GOAL

To enhance emotion detection by integrating Spotify's API, Emotionfy recommends playlists based on user emotions. The deep learning model analyzes facial expressions and maps them to Spotify playlists, offering a seamless blend of technology and music personalization.

MODEL ARCHITECTURE

Base Model: Trained a CNN on the FER Kaggle dataset for emotion classification.
Emotion-to-Playlist Mapping: A custom layer maps detected emotions to Spotify playlists for seamless recommendations.

TECH STACK

Frontend

React: Interactive user interface

Backend

Flask: API communication and emotion detection
Spotify Web API: Playlist recommendations

Data Processing

Data augmentation and preprocessing techniques for improved model generalization

Libraries and Tools

TensorFlow and Keras: Deep learning frameworks
OpenCV: Image processing
Pandas and Numpy: Data handling
Matplotlib and Seaborn: Data visualization
Jupyter Notebook: Model development and experiments

Integration

Spotify OAuth 2.0: Secure user authentication for playlist access
Spotify Web Playback SDK: Music playback within the interface

FEATURES

1. Emotion Detection
Analyze facial expressions in images or real-time video streams.
Detect and classify emotions such as Happy, Sad, Angry, etc.

2. Playlist Recommendation
Fetch Spotify playlists tailored to the detected emotion.
Play music directly within the app.

3. User Interface
Upload images or stream video for real-time emotion analysis.
Display detected emotions with an accompanying playlist.

EVALUATION METRICS

Emotion Detection: Accuracy, precision, recall, F1-score
Playlist Recommendation: Relevance and alignment based on user feedback

SETUP INSTRUCTIONS

Requirements

Python 3.8 or above
Spotify Developer Account (for API integration)

Steps

1. Clone the repository:

 git clone https://github.com/your-username/emotionfy.git cd server 

2. Install dependencies:

 pip install -r requirements.txt 

3. Configure Spotify API:
Create a Spotify Developer Account.
Set up an app and note the Client ID and Client Secret.
Update the config.py file with these credentials.

4. Run the application:

 npm dev run 

5. Run the server:

 python3 main.py 

USER INTERFACE

FUTURE ENHANCEMENTS

User Customization: Link personal Spotify accounts for tailored recommendations.
Platform Expansion: Integration with Apple Music and other music platforms.
Improved Emotion Detection: Incorporate video-based emotion tracking for dynamic recommendations.

CODE LICENSE

 MIT License
Copyright (c) 2024 Emotionfy Team
Permission is hereby granted, free of charge, to any person obtaining a copy of this software and associated documentation files (the "Software"), to deal in the Software without restriction, including without limitation the rights to use, copy, modify, merge, publish, distribute, sublicense, and/or sell copies of the Software, and to permit persons to whom the Software is furnished to do so, subject to the following conditions:
The above copyright notice and this permission notice shall be included in all copies or substantial portions of the Software.
THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE.