Optimizing feature extraction for speech recognition | IEEE Journals & Magazine | IEEE Xplore