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

Abstract

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, and BaggedSupport Vector Machine (SVM). The results for Bagged SVM were better than the other twoclassifiers. Top three individual features with highest accuracies were identified for each isolatedArabic consonant. To validate the results, a modified form of Sequential Floating Forward Selection(SFFS) process was used. Results showed that MFCC along with its first and second derivatives,energy, spectrum, and zero-crossing were the most suitable acoustic features for Arabicmispronunciation detection system. The proposed approach provided an average accuracy of 94.9%which was better than the previous best 92.95% for Arabic consonants.