Spectral_Smoothing

This exercise smooths the log magnitude spectrum of a frame of voiced speech using liftering

This MATLAB exercise processes a designated frame of voiced speech from a user specified speech file, using a designated window and calculates the log magnitude spectrum of the STFT of the windowed frame of speech. Next the exercise does LPC analysis of the speech frame using a pth-order predictor and computes the log magnitude spectrum of the LPC polynomial. Finally cepstral analysis is performed with the cepstrum liftered using a low frequency lifter (with lowest retained quefrency specified by the user). The exercise plots the three log magnitude spectrums on a common plot for comparison of the three log magnitude spectrums. The file '3.22 Spectral Smoothing.pdf' provides a User's Guide for this exercise.