Speech transformations based on a sinusoidal representation | IEEE Journals & Magazine | IEEE Xplore