Text

PENGENALAN EMOSI BERDASARKAN SUARA MENGGUNAKAN METODE SUPPORT VECTOR MACHINE (SVM)

---

Penilaian

---

Speaking is not only used to convey information but can also show a person's emotional state. Emotion recognition aims to help the machine to recognize emotions based on speech patterns issued by a person. Some of the problems found are how to find effective voice features and build a good model for emotion recognition. This study aims to identify seven types of emotions, namely anger, disgust, fear, happy, surprise, sad, and neutral by using the Support Vector Machine (SVM) classification method and Mel-Frequency Cepstral Coefficient (MFCC) feature extraction. The SVM method was chosen because this method is often used in classification cases and gives good performance results. In its implementation, voice data is subjected to a feature extraction process using the MFCC method and followed by a classification process using SVM. The test cases in this study are divided into three scenarios, namely the first scenario (70% training data and 30% test data), the second scenario (50% training data and 50% test data) and the third scenario is divided based on speaker's age, the aged 64 as training, and aged 26 as testing. Each Scenario was tested using several SVM kernels including linear, polynomial, and Radial Basis Function (RBF). Based on the test results in scenario one, the linear kernel gets an accuracy of 97.02% while in scenario two, the linear kernel gets an accuracy of 96.64%. But testing in the third scenario, the linear kernel gets an accuracy of 40.0%. This is because the features distribution in the data shows a different pattern from what happened in the first and second scenarios which are shown in the analysis by using Principal Component Analysis (PCA).

---

Availability

Inventory Code	Barcode	Call Number	Location	Status
2307000930	T88094	T880942023	Central Library (Referens)	Available but not for loan - Not for Loan

Detail Information

Series Title

-

Call Number

T880942023

Publisher

Inderalaya : ., 2022

Collation

xix, 104 hlm.; ilus.; 29 cm

Language

Indonesia

ISBN/ISSN

-

Classification

006.307

Content Type

-

Media Type

-

Carrier Type

-

Edition

-

Subject(s)

Sistem Pakar
Jurusan Teknik Informatika

Specific Detail Info

-

Statement of Responsibility

SEPTA

Other version/related

File Attachment

Comments

---

You must be logged in to post a comment