SOC Press, Indonesia Symposium on Computing (IndoSC) 2016

Font Size: 
SPEECH RECOGNITION ALGORITHM OF HIJAIYAH LETTER WITH PUNCTUATION USING LINEAR PREDICTIVE CODING (LPC) AND HIDDEN MARKOV MODEL (HMM)
Haby Bagus Prasetyo


Full Text: PDF

Abstract


Hijaiyah letter is the letter of constituent words in the Qur'an. Hijaiyah letter consists of 28 letters, with the letters symbolize consonant to vowel sounds while denoted by harokat / punctuation. Hijaiyah letter is part of the Arabic language that has characteristics both in writing and speech. Speech recognition system or voice recognition system is a system used to process voice signals into data that can be recognized by the computer [1]. To be able to do the voice recognition feature extraction methods are needed (feature extraction) and classifier. Sound signal that has been extracted character then generate information that can be analyzed for each variation existing voice signal. Of the characteristics that exist in each phoneme try to recognize it and convert it into text [2]. Methods of extraction of the characteristics used in this final project is Linear Predicitive Coding (LPC), then the feature generated from LPC quantized of each vector using the K-Means Clustering and for the classifier used when training and testing is the method of Hidden Markov Model (HMM). After several test scenarios obtained the best accuracy for testing is 58.93% and training is 99.60% with data 28 class.


Reference


[1] Fauzi, R. M., & Maharani, W. (2016). The Recognition of Hijaiyah Letter Pronunciation Using Mel Frequency Cepstral Coefficients and Hidden Markov Model. Advanced Science Letters, 22(8), 2043-2046. Fauzi, R. M., & Maharani, W. (2016). The Recognition of Hijaiyah Letter Pronunciation Using Mel Frequency Cepstral Coefficients and Hidden Markov Model. Advanced Science Letters, 22(8), 2043-2046.
Crossref

[2] Hakim¹, F. H., & Suyanto, I. I. T. PENGENALAN SINYAL SUARA PADA SPEECH-TO-TEXT MENGGUNAKAN LINEAR PREDICTIVE CODING (LPC) DAN HIDDEN MARKOV MODEL (HMM).

[3] Juang, B. H., & Rabiner, L. R. (1991). Hidden Markov models for speech recognition. Technometrics, 33(3), 251-272. Crossref

[4] Wisesty, U. N., & Liong, A. T. H. (2012, July). Indonesian speech recognition system using Discriminant Feature Extraction—Neural Predictive Coding (DFE-NPC) and Probabilistic Neural Network. In Computational Intelligence and Cybernetics (CyberneticsCom), 2012 IEEE International Conference on (pp. 158-162). IEEE.

[5] Rabiner, L. R., & Juang, B. H. (2006). Speech recognition: statistical methods. Encyclopedia of linguistics, 1-18. Crossref

[6] Shrawankar, U., & Thakare, V. M. (2013). Techniques for feature extraction in speech recognition system: A comparative study. arXiv preprint arXiv:1305.1145

[7] Yulita, I. N., Houw Liong The, & Adiwijaya. (2012). Fuzzy Hidden Markov Models for Indonesian Speech Classification. JACIII, 16(3), 381-387. Crossref

[8] Rabiner, L., & Juang, B. H. (1993). Fundamentals of speech recognition.

[9] Rahman¹, Y. H., & Adiwijaya, W. M. PENGENALAN HURUF HIJAIYAH BERBASIS SPEECH RECOGNITION MENGGUNAKAN HIDDEN MARKOV MODEL (HMM).

[10] Rabiner, L. R. (1989). A tutorial on hidden Markov models and selected applications in speech recognition. Proceedings of the IEEE, 77(2), 257-286. Crossref

[11] Wisesty, U. N., & Astuti, W. (2015, May). Feature extraction analysis on Indonesian speech recognition system. In Information and Communication Technology (ICoICT), 2015 3rd International Conference on (pp. 54-58). IEEE.

Last modified: 2016-11-02