Integrating Service Robots into Everyday Life Based on Audio Capabilities

Abstract

This paper presents a way in which TIAGo, a service robot from Pal Robotics, could be integrated in everyday life, based on his audio capabilities. We show that using 20 Mel frequency cepstral coefficients and 5-nearest neighbors we rich an overall correct classification rate of 99.27% in the testing phase. However, we do not recommend non-MFCC features for implementations with the TIAGo service robot, at least for the framework presented in the current paper.