PENGENALAN UCAPAN KATA AWAL PADA AYAT AL-QURAN MENGGUNAKAN DYNAMIC TIME WARPING

Nurridayanti Patmala, - (2018) PENGENALAN UCAPAN KATA AWAL PADA AYAT AL-QURAN MENGGUNAKAN DYNAMIC TIME WARPING. S1 thesis, Universitas Pendidikan Indonesia.

[img] Text
S_KOM_1300974_Title.pdf

Download (91kB)
[img] Text
S_KOM_1300974_Abstract.pdf

Download (249kB)
[img] Text
S_KOM_1300974_Table_of_content.pdf

Download (151kB)
[img] Text
S_KOM_1300974_Chapter1.pdf

Download (299kB)
[img] Text
S_KOM_1300974_Chapter2.pdf
Restricted to Staf Perpustakaan

Download (580kB)
[img] Text
S_KOM_1300974_Chapter3.pdf

Download (322kB)
[img] Text
S_KOM_1300974_Chapter4.pdf
Restricted to Staf Perpustakaan

Download (740kB)
[img] Text
S_KOM_1300974_Chapter5.pdf

Download (150kB)
[img] Text
S_KOM_1300974_Bibliography.pdf

Download (158kB)
[img] Text
S_KOM_1300974_Appendix.pdf
Restricted to Staf Perpustakaan

Download (603kB)
Official URL: http://repository.upi.edu

Abstract

Speech recognition mempunyai cakupan implementasi yang sangat luas dalam berbagai bidang kehidupan saat ini, seperti dalam sistem keamanan, alat medis, hingga bidang pendidikan, tak terkecuali dalam bidang pembelajaran keagamaan. Teknologi speech recognition dapat dijadikan alternatif media penunjang dalam proses menghafal Al-Quran. Penelitian ini akan mengimplementasikan algoritma Dynamic Time Warping ke dalam sebuah sistem isolated speech recognition untuk mengenali kata awal pada ayat Al-Quran. Proses recognition mencakup feature extraction dan feature matching. Feature extraction dilakukan untuk mengambil fitur dari sinyal data suara menggunakan Mel Frequency Cepstral Coefficients, sedangkan feature matching dilakukan untuk mendapatkan hasil recognition berupa nilai cost fitur yang paling minimum menggunakan Dynamic Time Warping. Hasil eksperimen menggunakan 5 orang model dan 3 jenis template speech menunjukkan bahwa dynamic time warping dapat mengenali suara yang diucapkan oleh orang yang sama, sangat sensitif terhadap suara yang diucapkan oleh orang yang berbeda, dan orang yang sama memiliki kecenderungan untuk melafalkan ayat dengan variasi yang konsisten.---- Speech recognition has a very wide and large scope of implementation in today’s living, such as security system, medical tools, up to education sector, including religious education. Speech recognition technology can be a media alternative for supporting Al-Quran memorizing process. This research would like to implements Dynamic Time Warping algorithm into an isolated speech recognition system for recognizing first word of each ayah of Al-Quran. The recognition process including two main phase, feature extraction and feature matching. Feature extraction is taken for getting features from speech data using Mel Frequency Cepstral Coefficients, whereas feature matching aims for getting the most minimum feature cost recognition result using Dynamic Time Warping. Experiments were done using 5 speech models and 3 kind of templates, and give results that dynamic time warping could recognize voice spoken by same person, is sensitive to voice spoken by different person, and a same person has tendency to pronounce ayah of Al-Quran with a stable consistency.

Item Type: Thesis (S1)
Additional Information: No. Panggil: S KOM NUR p-2018 ; Pembimbing: I. Rani Megasari, II. Yaya Wihardi ; NIM: 1300974
Uncontrolled Keywords: Isolated Speech Recognition, Dynamic Time Warping, Mel Frequency Cepstral Coefficients, Metode Menghafal Al-Quran.
Subjects: B Philosophy. Psychology. Religion > BP Islam. Bahaism. Theosophy, etc
L Education > L Education (General)
Q Science > QA Mathematics > QA75 Electronic computers. Computer science
Divisions: Fakultas Pendidikan Matematika dan Ilmu Pengetahuan Alam > Program Studi Ilmu Komputer
Depositing User: Mr. Arif Rezkyana Nugraha
Date Deposited: 20 Jun 2019 02:22
Last Modified: 20 Jun 2019 02:22
URI: http://repository.upi.edu/id/eprint/35518

Actions (login required)

View Item View Item