Design of Elevator Auxiliary Control System Based on Speech Recognition

In order to prevent people from touching the floor buttons when riding in the elevator in response to the COVID-19 pandemic, we propose a design method of elevator auxiliary control system based on speech recognition. The hardware design part of the system includes voice module, photoelectric sensor module and main control circuit module. On the basis of hardware design, the process from recognizing the voice commands of passengers in the elevator car to transmitting the recognition results to the elevator control main system is realized through programming. With the goal of improving the accuracy of speech recognition, the differential feature parameter algorithm based on the Mel-frequency cepstral coefficients (MFCC) is used. Then, our experiments show that least mean square (LMS) adaptive filtering is an effective method for the noise reduction of the voice signal when processing passengers' commands.