Recognize some of Arabic phonemes by using Mamdani NeuroFuzzy Network

Authors

  • Saba Abdul-Wahed S. Basrah Univ., Collage Since, Computer Dept.

Keywords:

Mamdani neurofuzzy, Adaptive back-propagation algorithm, zero crossing, speech recognizing .

Abstract

Modified mamdani neurofuzzy network scheme has been proposed and presented in this paper to recognize some of Arabic phonemes. The number of connecting is reduced to be equal to the number of member ship function plus one. to training this network ,we used some of sound segments which contain from two or three phonemes and recorded from more than one person in different age and both sexes  in the same environment by used the same recorded programming (sound forge 5.0)    . The proposed technique has been testing by using different group from sound segments which recorded in the same environment and recorded program. The percentage of recognition reaches to 87% to all recorded phonemes  

Downloads

Download data is not yet available.

Downloads

Published

2017-08-10

How to Cite

Abdul-Wahed S., S. (2017). Recognize some of Arabic phonemes by using Mamdani NeuroFuzzy Network. Journal of Al-Qadisiyah for Computer Science and Mathematics, 7(1), 131–145. Retrieved from https://jqcsm.qu.edu.iq/index.php/journalcm/article/view/92

Issue

Section

Math Articles