The design of voice recognition controller via grey relational analysis

Abstract

For the current Artificial Intelligence integration technology, if researchers want to establish the communicative platform, the voice recognition system is the basis of it. In the intelligent system, the voice recognition system can transform voice into text, and then through database query for matching and establishing the machine, and make the systematic machine understand human beings' voice. Instead of keying input, the voice recognition system builds the good interaction between the human beings and the machine. Therefore, the paper aims to combine the voice recognition controller with the infrared emission features in the model wagon. By applying human beings' voice with the voice recognition system, capturing the features of Mel-scale Frequency Cepstral Coefficients. Next, the grey relational theory is used to find the similar voice feature to be the recognition feature. Finally, through the micro chip to emit the infrared in model wagon. By the experimental verification, the accuracy rate can reach 80%, which proves that the system is feasible and can be a great help toward practical design of the products.