The cancellation of human sounds using synthesized soundwaves

Abstract

The principle of destructive interference is applied to reduce or eliminate the spoken sound in order to increase speech privacy. This research is divided into two parts: the synthesis of sound from formula, using both analytical linguistics and physical approaches, and the actual cancellation of simple to complex sounds in the real world setting. The results suggested that in a real-world setting, up to 60% intensity reduction could be repeatedly detected. When the complexity of the sound is increased, difference in interference pattern could be detected between sounds in frequency range within the human hearing spectrum. Furthermore, it is found that the generation of negative sound using the four formants is not comprehensive enough to interfere effectively with actual sound, therefore the spectral slice is needed to synthesize more complex sounds. After testing various variations, it is found that using 11 most intense frequency components adjusted in proportion according to each original amplitude best cancel the widest range of human sound samples. This method was used to test the effectiveness of actual human sound cancellation. Though effective cancellation was detected in the high frequencies range of the sound spectrum, other frequency ranges were barely affected. The difference between cancellation patterns of each frequency range might need investigating before the cancellation of human sound can be effectively applied in real life.