Emotion Based Music Recommendation System 

Recommend music based upon user's mood

Project Description 

This project focuses on developing an Emotion-Based Music Recommendation System that analyzes a user’s facial expressions to determine their current emotional state and recommends suitable music accordingly. A deep learning model based on the ResNet architecture is used to recognize emotions from facial images. The system then suggests the top 5 music tracks designed to uplift or match the user’s mood. The application is deployed as a web-based platform, where the frontend and backend communicate through REST APIs for seamless user interaction.

Features 

• Real-time facial emotion detection using deep learning

• ResNet-based CNN model for emotion classification

• Automatic recommendation of top 5 music tracks based on detected mood

• Web-based interface for easy user interaction

• Backend deployment using Flask REST APIs

• Suitable for final-year projects and AI-based recommendation systems

Tech Stack

• Python

• TensorFlow (Deep Learning)

• Convolutional Neural Networks (CNN)

• Flask (Backend Framework)

• HTML, CSS, JavaScript (Frontend)

• NumPy and other supporting Python libraries

System Requirements

• Python 3.10+

• Anaconda

• No GPU is required

Deliverables 

• Jupyter Notebook

• Dataset

• Setup over video call 

• WhatsApp support