This file contains the main part of DSR front-end. Speech samples * are read from input waveform file fr a me by fr a me. Feature * extraction is performed for each fr a me by calling basic * functions of the basic FE function package (see F