Abstract:
n recent years, the significance of human-drone interaction in scientific research has grown substantially. When engaging with drones, humans undertake various responsibilities, which are contingent upon the drone’s application and level of autonomy. This study aims to regulate the movements of unmanned aerial vehicles (UAVs) by utilizing a keyword spotting system. To achieve this, a deep neural network (DNN) is trained to comprehend user speech in both noisy and noiseless environments and generate the desired control commands accordingly. The hardware implementation of the developed system demonstrates both high accuracy in speech recognition and ease of control.
