Abstract: Creating voice signal features is an important and vital task, these features can be used in any secu-rity system to identify human and to identify the spoken word (phrase) by this human. In this research paper we will investigate three methods, each of them can be used to create a voice signal fea-tures, we will adopt and modify LBP method to be capable to create a features for a voice signal. The three me-thods will be programmed and implemented, the obtained experimental results will studied and analyzed in or-der to give some judgment about these method..
Keywords: Crest factor, dynamic range, Mu, sigma, LBP, CSLBP, MLBP, K-means, throughput, speed up.
[1]. R. Szabo, A. Gontean, I. Lie, "Sound Based Coin Recognition and Clapper", 16th International Conference on Soft Computing (MENDEL), pp. 509–516, June 23-25, 2010.
[2]. A. K. Paul, D. Das, M. M. Kamal, "Bangla Speech Recognition System Using LPC and ANN", Seventh International Conference on Advances in Pattern Recognition (ICAPR), Kolkata, India, pp. 171– 174, February 4-9, 2009.
[3]. R. Szabo, A. Gontean, "Human Voice Signal Synthesis and Coding", IFAC Proceedings Volumes, Vol. 46, No. 28, pp. 336-341, 2013.
[4]. K. Μ. Matrouk, A. Alhasanat, H. Alasha'ary, Z. Alqadi, H. M. AlShalabi, "Speech Fingerprint to Identify Isolated Word-Person", World Applied Sciences Journal, Vol. 31, No. 10, pp. 1767-1771, 2014. [5]. Jihad Nader Ismail Shayeb, Ziad Alqadi, Analysis of digital voice features extraction methods, International Journal of Educational Research and Development, v. 1, issue 4, pp. 49-55, 2019.