Room volume estimation based on statistical properties of binaural signals using humanoid robot

Abstract

Humans can recognize approximate room size only by binaural audition. However, sound reverberation is not negligible in most environments. The reverberation causes temporal fluctuation in the short-term Interaural Phase Differences (IPDs) of sound pressure. In this study, a novel method for estimating room volume is proposed based on the short-term IPDs of sound pressure by using a binaural humanoid robot. The humanoid robot turns its head toward a sound source, recognizes the sound source, then estimates the ego-centric distance by its stereovision. By interpolating the relations between room volume, average standard deviation and egocentric distance experimentally obtained for various rooms in a prepared database, the room volume was estimated by binaural audition from the average standard deviation of short-term IPDs at the estimated distance.