Abstract
A system for recognizing steganographic-transformed voice commands for controlling an unmanned aerial vehicle based on cepstral analysis has been developed, which allows efficient recognition and covert transmission of commands to an unmanned aerial vehicle by converting voice control commands into a kind of steganographic feature vector, which implies information hiding voice control of an unmanned aerial vehicle. A mathematical model of the algorithm for calculating small-frequency cepstral coefficients and a classifier for recognizing voice control commands has been synthesized to solve the problem of semantic identification and ensure the secrecy of control information of an unmanned aerial vehicle in a communication channel. The detailed results of preliminary experimental studies of the developed voice command recognition system and the computational algorithm of mel-frequency cepstral coefficients in MATLAB in the case of the identification of control commands "up", "down", "right", "left" spoken by different subjects.
Published Version (
Free)
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call