Operating a Robot by Nonverbal Voice Expressed with Acoustic Features

Abstract

This paper proposes methods for operating a robot by nonverbal voice. These methods enable operators to operate multi-degrees of freedom simultaneously and operate a robot intuitively by nonverbal voice by associating the nonverbal voice, tongue position and the coordinate of the robot’s hand. The voice is defined by formants or Mel Frequency Cepstral Coefficients (MFCC). Formants and MFCC are acoustic features and they show the characteristics of the vocal tract such as the mouth and the tongue. We propose two methods. One is the method in which voice expressed with overlapped formants ranges are used to change variable about robots’ operation. This method enables operators to operate multi-degrees of freedom simultaneously by nonverbal voice. The other is the method that operators tongue positions are distinguished by nonverbal voice. These tongue positions correspond to the coordinate of the robot’s hand and it enables the operators to operate a robot intuitively. We found the feasibility of the methods through experiments of simple tasks. These methods can realize operating a robot intuitively in continuous values by voice and can be utilized for user-friendly system.