Hasan Basri Slamat (2020) Penilaian Otomatis Pengucapan Huruf Arab Dalam Belajar Bahasa Arab Menggunakan Deep Neural Network. S1 thesis, Universitas Muhammadiyah Yogyakarta.

	Text (Halaman Judul) Halaman Judul.pdf Download (2MB)
	Text (Lembar Pengesahan) Lembar Pengesahan.pdf Restricted to Registered users only Download (997kB)
	Text (Abstrak) Abstrak.pdf Restricted to Registered users only Download (121kB)
	Text (Bab I) Bab I.pdf Download (118kB)
	Text (Bab II) Bab II.pdf Restricted to Registered users only Download (444kB)
	Text (Bab III) Bab III.pdf Restricted to Registered users only Download (42kB)
	Text (Bab IV) Bab IV.pdf Restricted to Registered users only Download (602kB)
	Text (Bab V) Bab V.pdf Restricted to Registered users only Download (6kB)
	Text (Daftar Pustaka) Daftar Pustaka.pdf Restricted to Registered users only Download (126kB)
	Text (Lampiran) Lampiran.pdf Restricted to Registered users only Download (20kB)
	Text (Naskah Publikasi) Naskah Publikasi.pdf Restricted to Registered users only Download (269kB)
	Text (Full Text) Full Text.pdf Restricted to Repository staff only Download (5MB)

Abstract

Learning Al-Quran starts with learning the hijaiyah letters. Generally, Al-Qur'an learning is carried
out at TPA / TPQ. However, at the end of 2019, the whole world was faced with a pandemic which
caused all activities including learning the Koran to not be carried out as usual at TPA / TPQ. By
utilizingtechnology deep learning that is increasingly developing in the world. This study
usestechnology Deep Neural Network by utilizingtechniques Speech Recognition. Voice data used
in this study were 3640 voice data divided into two, which is 2800 for train data and 840 for test
data. To carry out an automatic assessment of the hijiaiyah letter sound data, training needs to be
done with some preprocessing where the final results will be compared to determine the best
preprocessing. The resulting process in this study using preprocessing simple Convolutional
Neural Network of 92,02%. This value is higher compared to using preprocessing Spectrogram
91,43% or preprocessing Mel Scale Spectrogram 87,86% and MFCC preprocessing is only
78.93%.

Dosen Pembimbing:	Asroni, S.T., M.Eng. and Cahya Damarjati, S.T., M.Eng. | NIDN0526047401, NIDN0515038702
Item Type:	Thesis (S1)
Divisions:	Fakultas Teknik > S1 Teknologi Informasi
Depositing User:	Unnamed user with email robi@umy.ac.id
Date Deposited:	12 Oct 2021 01:22
Last Modified:	05 Nov 2021 01:56
URI:	https://etd.umy.ac.id/id/eprint/2496

Actions (login required)

View Item