Speaker recognition using lower formants | IEEE Conference Publication | IEEE Xplore