Ekstraksi Fitur Pengenalan Emosi Berdasarkan Ucapan Menggunakan Linear Predictor Ceptral Coeffecient Dan Mel Frequency Cepstrum Coefficients
DOI:
https://doi.org/10.12928/mf.v1i2.1259Abstract
Ucapan suara memiliki informasi penting yang dapat diterima oleh otak melalui gelombang suara. Otak menerima gelombang suara melalui alat pendengaran dan menghasilkan suatu informasi berupa pesan, bahasa, dan emosi. Pengenalan emosi wicara merupakan teknologi yang dirancang untuk mengidentifikasi keadaan emosi seseorang dari sinyal ucapannya. Hal tersebut menarik untuk diteliti, karena berkaitan dengan teknologi zaman sekarang yaitu pada penggunaan smartphone di berbagai macam aktivitas sehari-hari. Penelitian ini membandingkan ekstraksi fitur Metode LPC dan Metode MFCC. Kedua metode ekstraksi tersebut diklasifikasi menggunakan Metode Jaringan Syaraf Tiruan (MLP) untuk pengenalan emosi. Masing-masing metode menggunakan data emosi marah, bosan, bahagia, netral, dan sedih. Data dibagi menjadi dua, yaitu data testing dan data data training dengan perbandingan 80:20. Arsitektur jaringan yang digunakan adalah tiga lapisan yaitu lapisan input, lapisan tersembunyi, dan lapisan output. Parameter MLP yang digunakan learning rate = 0.0001, epsilon = 1e-08, epoch = 500, dan Cross Validation = 5. Hasil akurasi pengenalan emosi dengan ekstraksi fitur LPC sebesar adalah 28%. Sedangkan hasil akurasi dengan ekstraksi fitur MFCC sebesar 61,33%. Hasil akurasi ini bisa ditingkatkan dengan menambahkan data yang lebih banyak lagi, terutama untuk data testing. Perlunya pengujian pada nilai parameter jaringan MLP, yaitu dengan mengubah nilai-nilai parameter, karena dapat mempengaruhi tingkat akurasi pengenalan. Selain itu penentuan ekstraksi fitur dan klasifikasi metode yang lain juga dapat digunakan untuk mencari nilai akurasi pengenalan emosi yang lebih baik lagi.
References
S. Vaishnav and S. Mitra, "Speech Emotion Recognition: A Review," Int. Res. J. Eng. Technol. IRJET, vol. 3, no. 04, pp. 313-316, 2016.
S. Lalitha, A. Madhavan, B. Bhushan, and S. Saketh, "Speech Emotion Recognition," 2014 Int. Conf. Adv. Electron. Comput. Commun. ICAECC 2014, vol. 7, 2015, doi: 10.1109/ICAECC.2014.7002390.
A. B. Gumelar, "Human Voice Emotion Identification Using Prosodic and Spectral Feature Extraction Based on Deep Neural Networks," 2019 IEEE 7th Int. Conf. Serious Games Appl. Health SeGAH, pp. 1-8, 2019.
M. D. Pell and S. A. Kotz, "On the time course of vocal emotion recognition," PLoS One, vol. 6, no. 11, p. e27256, 2011.
I. Idrisa, M. S. H. Salamb, and M. S. Sunarc, "Speech Emotion Classification Using SVM and MLP on Prosodic and Voice Quality Features," J. Teknol., vol. 78, 2015.
G. Liu, W. He, and B. Jin, "Feature Fusion of Speech Emotion Recognition Based Deep Learning," in 2018 International Conference on Network Infrastructure and Digital Content (IC-NIDC), 2018, pp. 193-197, doi: 10.1109/ICNIDC.2018.8525706.
Y. Sun, G. Wen, and J. Wang, "Weighted spectral features based on local Hu moments for speech emotion recognition," Biomed. Signal Process. Control, vol. 18, pp. 80-90, 2015, doi: 10.1016/j.bspc.2014.10.008.
S. S. Swaminathan and J. Thangaiyan, "Emotion Speech Recognition using MFCC and Residual Phase in Artificial Neural Network," Int. J. Eng. Res. Sci. Technol., vol. 4 No.3, no. August, pp. 106-113, 2015.
Irmawan, H. Hikmarika, D. W. Sari, and M. C. Tammimi, "Pengenalan Kata dengan Metode Linear Predictive Coding dan Jaringan Syaraf Tiruan Pada Mobile Robot," in Conference on Information Technology and Electrical Engineering, 2014, no. October 2014, pp. 139-144.
K. V. Krishna Kishore and P. Krishna Satish, "Emotion Recognition in Speech using MFCC and Wavelet Features," Proc. 2013 3rd IEEE Int. Adv. Comput. Conf. IACC 2013, pp. 842-847, 2013, doi: 10.1109/IAdCC.2013.6514336.
Downloads
Published
Issue
Section
License
Copyright (c) 2019 Siti Helmiyah, Imam Riadi, Rusydi Umar, Abdullah Hanif
This work is licensed under a Creative Commons Attribution-ShareAlike 4.0 International License.
Start from 2019 issues, authors who publish with JURNAL MOBILE AND FORENSICS agree to the following terms:
- Authors retain copyright and grant the journal right of first publication with the work simultaneously licensed under a Creative Commons Attribution License (CC BY-SA 4.0) that allows others to share the work with an acknowledgment of the work's authorship and initial publication in this journal.
- Authors are able to enter into separate, additional contractual arrangements for the non-exclusive distribution of the journal's published version of the work (e.g., post it to an institutional repository or publish it in a book), with an acknowledgment of its initial publication in this journal.
- Authors are permitted and encouraged to post their work online (e.g., in institutional repositories or on their website) prior to and during the submission process, as it can lead to productive exchanges, as well as earlier and greater citation of published work.
This work is licensed under a Creative Commons Attribution-ShareAlike 4.0 International License.