Enhancement of a text-independent speaker verification system by using feature combination and parallel structure classifiers | Publicación