Sound source localization with the aid of excitation source information in home robot environments

This paper is concerned with multiple microphone-based sound source localization in home robot environments. For this purpose, we use the excitation source information to determine the time-delay between each two microphones from speech source when robot's name is called. Furthermore, we present a novel method to estimate the reliable localization angle from the obtained time-delay values. Thus, it can be used as a core technique in conjunction with human-robot interaction that can naturally interact between human and robots in home robot applications. The experimental results on sound localization database revealed that the presented method showed a good performance in comparison with the conventional methods such as time difference of arrival (TDOA) and generalized cross correlation- phase transform (GCC-PHAT) in distant-varying, noise and reverberant environments.