FEATURE SELECTION FOR ARABIC MISPRONUNCIATION DETECTION BASED ON SEQUENTIAL FLOATING FORWARD SELECTION AND DATA MINING CLASSIFIERS

Feature selection process is used to reduce the feature vector length and identify thediscriminative features. Many acoustic-phonetic features including Mel-Frequency CepstralCoefficient (MFCC), Energy, Pitch, Zero-crossing, spectrum were tested individually for Arabicmispronunciation detection using three classifiers; Random Forest, Bayesian classifier, BaggedSupport Vector Machine (SVM). The results Bagged SVM better than other twoclassifiers. Top individual with highest accuracies identified each isolatedArabic consonant. To validate results, a modified form of Sequential Floating Forward Selection(SFFS) was used. Results showed that MFCC along its first second derivatives,energy, spectrum, zero-crossing most suitable acoustic system. proposed approach provided an average accuracy 94.9%which previous best 92.95% Arabic consonants.