The phonetic recognition of arabic figures using neural network

Speech recognition is a knowledge based process for automaticallyextracting various information from speech ultterd by human beings.Speech recognition technology has made steady process in its 40-yearshistory and has succeeded in creating several substantial applications. Thegoal of speech recognition research is to produce a machine which willrecognize accurately normal human speech from any speaker.One of the most important achievements for the feature extraction partand speech recognition techniques is the linear predictive coding (LPC).In this work LPC is used for extracting the formants of the Arabic digitsfrom (0 - 9) and DTW techniques is used for time normalization.The artificial neural networks are used for speech recognition as aclassifier for two main systems which are dependent system andindependent system for speakers.The recognition accuracy for dependent system is 98.8% but therecognition accuracy for independent system for males only is 88% and therecognition accuracy for females only is 87.8%. When combined groups areused, the recognition accuracy is 78.8% so a new system is used forimproving this result after dividing the Arabic digits into groups andprepared as inputs to the system where three neural networks are used toclassify the Arabic digits. The system improve the recognition accuracy into95.5%.