SELECTION OF DISCRIMINATIVE FEATURES FOR ARABIC PHONEME’S MISPRONUNCIATION DETECTION

Pronunciation training is an important part of Computer Assisted Training (CAPT) systems. Mispronunciation detection systems recognized pronunciation mistakes from user’s speech and provided them feedback about their pronunciation. Acoustic phonetic features plays a vital role in classification based applications. This research work investigated the suitability various acoustic features: pitch, energy, spectrum flux, zero-crossing, Entropy MelFrequency Cepstral Coefficients (MFCCs). Sequential Forward Selection (SFS) was used to find out most suitable computed feature set. study K-Nearest Neighbors (K-NN) classifier detect Arabic phonemes. selected set discriminative for each phoneme. K-NN achieved accuracy 92.15% mispronunciation Phonemes.