Speech Steganalysis Using Delay Vector Variance Based Features

We investigate the use of delay vector variance-based features for recorded speech steganalysis. Considering  that data hiding within a speech signal distorts the properties of the original speech signal, we design a steganalyzer that uses surrogate data based delay vector variance ( DVV ) features to detect the existence of a stego-signal. We evaluate the proposed DVV features with other chaotic-type features separately and all together. Also the applicability of the proposed method to general audio is discussed.