Implementasi MIR (Music Information Retrieval) pada Modul Genre Recognition dan Deep Learning Classification untuk Aplikasi Musicmoo

Mahardika, Muhammad Nezar (2018) Implementasi MIR (Music Information Retrieval) pada Modul Genre Recognition dan Deep Learning Classification untuk Aplikasi Musicmoo. Undergraduate thesis, Institut Teknologi Sepuluh Nopember.

[img]
Preview
Text
5114100001-Undergraduate_Thesis.pdf - Accepted Version

Download (2MB) | Preview

Abstract

Saat ini, perkembangan dunia musik mengalami peningkatan. Peningkatan ini terjadi di dalam sisi industri musik maupun dalam sisi perkembangan genre musik itu sendiri. Dalam perjalanannya, genre musik mengalami perkembangan dengan terbentuknya bagian-bagian dalam sebuah genre, yaitu subgenre. Selain itu, perkembangan di industri musik juga membuat semakin banyaknya lagu-lagu baru bermunculan. Seringkali kita tidak mengetahui judul lagu-lagu tersebut. Metode pemanggilan kembali informasi suatu musik atau yang sering disebut Music Information Retrieval (MIR) bisa diterapkan untuk pendeteksian lagu dan prediksi subgenre pada suatu lagu. Langkah pertama yang dilakukan adalah melakukan ekstraksi fitur audio berbasis MPEG-7. Selanjutnya adalah menerapkan algoritma Bhattacharyya Distance untuk pendeteksian lagu dan Deep Neural Network (DNN) untuk klasifikasi suatu subgenre. Hasil akhirnya adalah dapat dilakukannya pendeteksian dan prediksi subgenre suatu lagu. ============= Currently, the development of the music world has increased. This increase occurs within the music industry as well as in the developmental side of the music genre itself. In its journey, the music genre has developed with the formation of parts in a genre, the subgenre. In addition, developments in the music industry also made more and more new songs emerging. Often we do not know the title of the songs. The method of calling back information of a music or often called Music Information Retrieval (MIR) can be applied to the detection of songs and subgenre predictions on a song. The first step is to extract the MPEG-7 based audio features. the second step is to apply Bhattacharyya Distance algorithm for song detection and Deep Neural Network (DNN) for the classification of a subgenre. The result is the detection and prediction of a subgenre of a song.

Item Type: Thesis (Undergraduate)
Additional Information: RSIf 006.32 Mah i
Uncontrolled Keywords: MIR; MPEG-7; subgenre; DNN; Bhattacharyya
Subjects: T Technology > TK Electrical engineering. Electronics Nuclear engineering > TK5105.546 Computer algorithms
Z Bibliography. Library Science. Information Resources > ZA Information resources > Z699.5 Information storage and retrieval systems
Divisions: Faculty of Information Technology > Informatics Engineering > 55201-(S1) Undergraduate Thesis
Depositing User: Mahardika Muhammad Nezar
Date Deposited: 26 Feb 2018 04:55
Last Modified: 22 Sep 2020 03:42
URI: https://repository.its.ac.id/id/eprint/50547

Actions (login required)

View Item View Item