Wicaksono, R. Mahendra Agung
(2025)
IMPLEMENTASI ALGORITMA SILENCE REMOVAL DAN ENDPOINT DETECTION UNTUK MENGHILANGKAN NOISE PADA APLIKASI MEDIA PEMBELAJARAN PRONUNCIATION BAHASA INGGRIS
(Studi Kasus: SMPN 1 Maleber).
S1 / D3 thesis, Universitas Kuningan.
Abstract
Penelitian ini bertujuan untuk mengembangkan aplikasi media pembelajaran pronunciation bahasa Inggris berbasis Android di SMPN 1 Maleber, menggunakan algoritma Silence Removal dan Endpoint Detection untuk deteksi keheningan dan mengurangi noise selama evaluasi pelafalan. Siswa mengalami kendala dalam mengucapkan kata-kata dengan baik dan benar, dan kegaduhan selama evaluasi membuat hasil pelafalan sulit terdengar jelas oleh guru.
Aplikasi ini menyajikan dua materi, yaitu conversation dan vocabulary, dengan fitur speech recognition untuk memvalidasi ucapan pelafalan siswa. Algoritma Silence Removal menghapus bagian sinyal audio yang tidak bersuara, sementara Endpoint Detection mendeteksi awal dan akhir sinyal suara untuk meningkatkan kejelasan.
Pengembangan aplikasi dilakukan menggunakan metode Rational Unified Process (RUP), dan pengujian dilakukan melalui metode black box dan White box serta User Acceptance Testing (UAT). Hasil UAT mendapatkan persentase sebesar 94% yang menunjukkan bahwa aplikasi dapat diterima dengan baik oleh pengguna. Guru dan siswa menyatakan aplikasi ini membantu dalam proses belajar mengajar dengan lebih interaktif dan efektif. Implementasi aplikasi ini diharapkan mampu meningkatkan keterampilan pronunciation siswa serta memperbaiki pengalaman belajar secara keseluruhan.
Kata Kunci : Pronunciation, Silence Removal, Endpoint Detection, Speech Recognition, Media Pembelajaran, Noise.
This study focuses on creating an Android-based application to support English pronunciation learning for students at SMPN 1 Maleber. The app utilizes Silence Removal and Endpoint Detection algorithms to help identify silent parts and reduce background noise during pronunciation assessments, addressing challenges students face with accurate word pronunciation. Background noise and interruptions during assessments can make it difficult for teachers to clearly evaluate students' pronunciation.
The app includes two core learning modules: conversation and vocabulary. It uses speech recognition to check the accuracy of students' pronunciation. The Silence Removal algorithm removes non-voiced portions of the audio, while the Endpoint Detection algorithm marks the start and end of spoken sounds, improving clarity.
Developed using the Rational Unified Process (RUP) methodology, the application was tested through black-box, white-box, and User Acceptance Testing (UAT). With a 94% acceptance rate in UAT, the results indicate high user satisfaction. Both teachers and students found the app valuable for creating a more interactive and effective learning experience. This application is anticipated to improve student’s pronunciation skills and enrich their overall learning journey.
Kata Kunci : Pronunciation, Silence Removal, Endpoint Detection, Speech Recognition, Learning Media, Noise.
Item Type: |
Thesis
(S1 / D3)
|
Uncontrolled Keywords: |
Pronunciation, Silence Removal, Endpoint Detection, Speech Recognition, Media Pembelajaran, Noise.
Pronunciation, Silence Removal, Endpoint Detection, Speech Recognition, Learning Media, Noise. |
Subjects: |
L Education > L Education (General) L Education > LB Theory and practice of education T Technology > T Technology (General) |
Divisions: |
Fakultas Ilmu Komputer > S1 Teknik Informatika |
Depositing User: |
S.Kom R. Mahendra Agung Wicaksono
|
Date Deposited: |
08 May 2025 02:38 |
Last Modified: |
08 May 2025 02:38 |
URI: |
https://rama.uniku.ac.id/id/eprint/2658 |
Actions (login required)

- View Item