Music Genre Classification

Ooi, Joo Teik (2025) Music Genre Classification. Final Year Project (Bachelor), Tunku Abdul Rahman University of Management and Technology.

Text RDS_Ooi Joo Teik_Fulltext.pdf Restricted to Registered users only Download (2MB)

Abstract

To meet the growing need for efficient organisation of large music libraries and improved recommendation systems on digital music platforms, this project will develop a music genre classification system using machine learning techniques to enhance user experience and help content discovery. This project aims to create a model that can accurately classify music into predefined genres by exploring traditional machine learning algorithms such as K-Nearest Neighbour (KNN) and Support Vector Machines (SVM), and advanced deep learning models such as Convolutional Neural Networks (CNN). Trained on audio features and spectrograms such as the Mel Spectrogram, the CNN model achieved an accuracy of 97%, and the performance is better than the traditional models. Furthermore, the scope of the project was not limited to machine learning model training but also included a fully functional web-based system where users can log in, upload music, play tracks, and classify music genres in real time. The system’s architecture integrates the model with front-end, back-end, and database components to ensure user-friendly interaction and efficient data management.