Analisis karakter suara penderita bibir sumbing terhadap suara normal menggunakan metod mel frequency cepstral coeffients dan dynamic time wraping

Anggoro, Akhmad (2015) Analisis karakter suara penderita bibir sumbing terhadap suara normal menggunakan metod mel frequency cepstral coeffients dan dynamic time wraping. Undergraduate thesis, Universitas Islam Negeri Maulana Malik Ibrahim.

Text (Fulltext)
10650089.pdf - Accepted Version
Available under License Creative Commons Attribution Non-commercial No Derivatives.

Download (3MB) | Preview



Suara merupakan sarana manusia dalam berkomunikasi secara verbal. Kelainan yang terjadi pada produksi suara dapat menghambat komunikasi seseorang karena komunikasi verbal merupakan komunikasi utama dalam kehidupan manusia. Sebagai contoh pada penderita bibir sumbing dan langit-langit, pola suara mereka tidak normal dan sering sulit untuk di mengerti. Oleh sebab itu, dengan memanfaatkan perkembangan teknologi saat ini, aplikasi identifikasi suara dibuat untuk membantu para penderita bibir sumbing agar dapat berkomunikasi verbal secara efektif.
Saat ini pengolahan suara banyak digeluti sebagai sarana penelitian individu maupun kelompok karena pengolahan suara menarik untuk diterapkan. Software ini memiliki 2 sisi, yakni sisi suara uji dan sisi suara pembanding, sehingga akan di ketahui apakah metode MFCC dapat mengektraksi kedua suara tersebut. Lalu kedua suara itu akan di bandingkan kemiripannya menggunakan DTW.

Berdasarkan hasil uji coba yang telah dilakukan, aplikasi mampu membandingkan 2 suara meskipun terdapat beberapa eror yang disebabkan pada kedua suara memiliki perbedaan kecepatan penyebutan. Dalam pembuatan software ini, perlu di perhatikan metode pengektraksi suara yang digunakan, yakni mel frequency cepstral coeffients (MFCC), dan variable suara perlu penelitian lebih lanjut karena penderita bibir sumbing memiliki suara yang berbeda dari suara normal.


Lately, sound processing became more involved in the field for a means of individual and groups recearches. Because sound processing is very interesting to be applied. The continued development of the times, evolving technology anyway. Every man thinks practically, simple, efficient and easy. And the sound is one of a vital human communication. But what if the sound was impaired. Like case patients with cleft lip and palate. Their voice is not the normal pattern and often difficult to understand. Therefore, utilize existing technology in sound identification applications aim at people with cleft lip. So that not only expert can understand but laymen and patients themselves can know his voice.

In making this application there are things that need to be noticed that the sound extraction method, namely mel frequency cepstral coeffients. And sound variable need further research. Because the sound of cleft lip patients have a different sound than normal sound. And to involve the normal sound for comparison. This application has 2 sides. Those are test sound side and comparison sound side. Where will be known whether the MFCC method can extract two voices. Then both sounds would be compared the resemblance using DTW. Based on the results of trials that have been done. Applications are able to compare the two sounds. Although there are some errors because both the voice has a mention of the difference. There are faster and slower. And hopefully from these applications could be developed better. Because this application can help a lot of people, especially for Patients with cleft lip.

Item Type: Thesis (Undergraduate)
Supervisor: Chamidy, Totok and Nurhayati, Hani
Keywords: engolahan Suara; Bibir Sumbing; Sound Processing; Cleft Lip; Mel Frequency Cepstral; Dynamic Time Wraping
Departement: Fakultas Sains dan Teknologi > Jurusan Teknik Informatika
Depositing User: Dian Anesti
Date Deposited: 13 Oct 2017 07:10
Last Modified: 13 Oct 2017 07:10

Actions (login required)

View Item View Item