Front-Back Confusion Resolution in 3D Sound Localization with HRTF Databases

Sound localization plays an important role in everyday life of a human for as long as people have lived on the Earth. It helps us to understand and separate sounds coming from different sound sources and to catch necessary information from the acoustical world. Several cues for localizing a position of a sound source, such as ITD, HRTF, recorded by dummy head microphones, are investigated in this paper. We used signals which were considered to be basic signals of an imaginary round-shaped room. Our goal of the research was to localize random signals produced from different positions using the information about basic signals. One of the main issues during sound localization was the problem of front-back confusion, which was successfully resolved by using reference signals. In this paper we propose sound localization algorithm with front-back confusion resolution.