Voice pathology analysis has been one of the useful tools in the diagnosis of the pathological voice. This method is non-invasive, inexpensive and reduces time required for analysis. This paper investigates the feature extraction based on the Dual-Tree Complex Wavelet Packet Transform (DT-CWPT) with entropies and energy measures tested with two classifiers, k-Nearest Neighbors (k-NN) and Support Vector Machine (SVM). Feature selection using ReliefF algorithm is applied to reduce redundancy features set and obtain the optimum features for classification. Massachusetts Eye and Ear Infirmary (MEEI) voice disorders database and Saarbruecken Voice Database (SVD) are used. This research was done on multiclass and by specific pathology. The experimental results automates the process of voice analysis hence produce promising results of the presence of diseases in vocal folds.