Transkripsi Alat Musik Demung Berbasis STFT Menggunakan Deep Learning

Hermawan, Andi Rokhman (2022) Transkripsi Alat Musik Demung Berbasis STFT Menggunakan Deep Learning. Masters thesis, Institut Teknologi Sepuluh Nopember.

Text 07111850050003-Master_Thesis.pdf - Accepted Version Restricted to Repository staff only until 1 October 2024. Download (8MB) | Request a copy

Abstract

Belajar memainkan gamelan akan lebih mudah jika terdapat petunjuk notasi.Proses konversi dari sebuah sinyal musik menjadi notasi dinamakan transkripsi.Transkripsi pada salah satu musik gamelan seperti Demung dapat dilakukanmenggunakan metode Deep Learning. Tiap notasi Demung dari Demung 6-rendah hinggal 1-tinggi dikonversi ke domain waktu-frekuensi menggunakanSTFTShort-Time Fourier Transform.Kemudian data domain waktu-frekuensi tersebut akan digunakan sebagai input dari multilayer perceptron.Metode training yang yang digunakan adalahsingle labeldari tiap notasi.Nilai keluaran dari model adalah sebuah notasi hasil transkripsi. Akurasimodel dalam penelitian ini setelah diuji dengan sampel sinyal audio DemungSlendro adalah 98.375%.
===========================================================================================================================
Learning to play a gamelan instrument would be easier when there’s a musicalnotation guide. The process of converting a musical signal into a notationguide is called transcription. We would like to transcript the gamelan musicespecially the Demung instrument using the Deep Learning method. EachDemung’s note from 6-low until 1-high would be converted to the time-frequency domain using STFT (Short-Time Fourier Transform). Then, thosedata will be treated as an input for the multilayers perceptron. The trainingmethod is a single label of each notation. The output returned by the modelis a music roll transcription. The accuracy of the model is around 98.375%.

Item Type:	Thesis (Masters)
Additional Information:	RTE 621.389 3 Her t-1 2022
Uncontrolled Keywords:	Gamelan, Pengenalan Notasi, Deep-Learning, Transkripsi Musik Otomatis, Notation Recognition, Automatic Music Transcription
Subjects:	Q Science > QC Physics > QC221 Acoustics. Sound R Medicine > R Medicine (General) > R858 Deep Learning
Divisions:	Faculty of Intelligent Electrical and Informatics Technology (ELECTICS) > Electrical Engineering > 20101-(S2) Master Thesis
Depositing User:	Anis Wulandari
Date Deposited:	09 Nov 2022 04:21
Last Modified:	09 Nov 2022 04:21
URI:	http://repository.its.ac.id/id/eprint/95073

Actions (login required)

View Item