KLASIFIKASI GENRE MUSIK BERDASARKAN COVER ALBUM MENGGUNAKAN METODE DEEP LEARNING

VALENDRIL, GENTA AGSAL and Supardi, Julian and Arsalan, Osvari (2023) KLASIFIKASI GENRE MUSIK BERDASARKAN COVER ALBUM MENGGUNAKAN METODE DEEP LEARNING. Undergraduate thesis, Sriwijaya University.

	Text RAMA_55201_09021281823049.pdf - Accepted Version Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (5MB) | Request a copy
	Text RAMA_55201_09021281823049_TURNITIN.pdf - Accepted Version Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (3MB) | Request a copy
	Text RAMA_55201_09021281823049_0010077210_0028068806_01_front_ref.pdf - Accepted Version Available under License Creative Commons Public Domain Dedication. Download (5MB)
	Text RAMA_55201_09021281823049_0010077210_0028068806_02.pdf - Accepted Version Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (5MB) | Request a copy
	Text RAMA_55201_09021281823049_0010077210_0028068806_03.pdf - Accepted Version Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (5MB) | Request a copy
	Text RAMA_55201_09021281823049_0010077210_0028068806_04.pdf - Accepted Version Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (5MB) | Request a copy
	Text RAMA_55201_09021281823049_0010077210_0028068806_05.pdf - Accepted Version Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (5MB) | Request a copy
	Text RAMA_55201_09021281823049_0010077210_0028068806_06.pdf - Accepted Version Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (10kB) | Request a copy
	Text RAMA_55201_09021281823049_0010077210_0028068806_06_ref.pdf - Bibliography Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (140kB) | Request a copy
	Text RAMA_55201_09021281823049_0010077210_0028068806_07_lamp.pdf - Accepted Version Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (8kB) | Request a copy

Abstract

Classification of music genres has been carried out using various features. Research by Zhang (2022) conducted music genre classification based on sound frequency waves, resulting in an accuracy of 91% with a dataset of 1000 audio files. Music genre classification based on lyrics has also been studied by Oramas et al. (2017), the research used a dataset of 31,471. This study focuses on solving the problem of music genre classification using a dataset of music album covers with VGG-16. VGG-16 itself is a deep learning model developed by Simonyan & Zisserman (2015) as part of the ILSVRC-2014 competition. By using VGG-16 and a dataset consisting of music album covers from 7 genres of music, the study was conducted with 30 epochs. The trained VGG-16 model was then tested using a separate dataset for testing purposes. Initially, music genres were classified into 19 labels, with labels having less than 20% representation. Subsequently, adjustments were made to reduce the labels to 7, with a configuration of 30 epochs. The testing results of this study showed an accuracy of 56%. The accuracy improvement occurred with the adjustment of music genres by 166%. It can be concluded that the number of music genres determines the success rate of the VGG-16 method in classification. Keywords: music genre classification, VGG-16, deep learning, album covers, Convolutional Neural Networks (CNNs)

Item Type:	Thesis (Undergraduate)
Uncontrolled Keywords:	klasifikasi genre musik, VGG-16, Deep Learning, cover album, Convolutional neural networks (CNNs)
Subjects:	M Music and Books on Music > M Music Q Science > Q Science (General) > Q334-342 Computer science. Artificial intelligence. Algorithms. Robotics. Automation.
Divisions:	09-Faculty of Computer Science > 55201-Informatics (S1)
Depositing User:	Genta Agsal Valendril
Date Deposited:	28 May 2024 14:02
Last Modified:	28 May 2024 14:02
URI:	http://repository.unsri.ac.id/id/eprint/145878

Actions (login required)

View Item