International Journal of Computer Applications |
Foundation of Computer Science (FCS), NY, USA |
Volume 81 - Number 4 |
Year of Publication: 2013 |
Authors: Zaid Y. Mohammed, Abdul Sattar M. Khidhir |
10.5120/14003-2048 |
Zaid Y. Mohammed, Abdul Sattar M. Khidhir . Real-Time Arabic Speech Recognition. International Journal of Computer Applications. 81, 4 ( November 2013), 43-45. DOI=10.5120/14003-2048
Speech recognition system needs to perform a high complex calculation and short time to complete it. This is a big challenge for the real-time systems. However, using a simple and fast algorithm may do this task for the slow systems. Thus, the main objective of this paper is to design and implement a Real-Time Arabic Speech Recognition system using MATLAB environment. It is capable of accurately identifying some letters while remaining simple and fast. It uses the Mel-Frequency Cepstral Coefficients (MFCCs) as a feature extraction and Euclidean distance to compare the test sound and the database. A recognition rate of 89. 6% has been reached.