Simultaneous Speech Detection Based On Mfcc-Dtw With Two-Stage Normalization

In Air Traffic Control a serious safety risk is represented by undetected simultaneous transmissions from different airplanes. In this paper, we approach this issue through a speech analysis algorithm, which combines traditional Mel Frequency Cepstral Coefficients extraction, a new two-stage normalization and widely used Dynamic Time Warping. In this way, we were able to extend the simultaneous speech detection capability in Voice Communication Systems of Air Traffic Control.