الفهرس 
acknowledgmentOnly 14 pages are availabe for public view
 |  | from | 105 |  |  |
| | | from | 105 | | |
Abstract
Automatic Speech Recognition (ASR) is the technology that allows a computer to identify the words that a person speaks into a microphone or telephone; it is a very popular research goal in the field of machine intelligence specially in Human-machine interaction.
The present research addresses developing a cognitive Arabic speech recognition application. The developed system used to enhance the speech abilities for Pre-school children, by recognizing the child’s failure speech and provide a cognitive multimedia program which helps him to repeat the word several times till pronounces the correct one.
The present research consists of two parts, Arabic speech recognition system and cognitive multi -media system: The Arabic speech recognition system starts with a pre-processing stage which begins with sampling the speech signals and determining the speech boundaries. As a second stage the features are extracted by using the Mel-Frequency Cepstral Coefficients (MFCC) technique. The extracted features are then reduced by using principal component analysis (PCA). The third stage uses artificial neural network (ANN) for recognition.
The cognitive system contains a cognitive content represented in an attractive multi-media program which is supported with pictures, songs, practices and some stories.