The famous statement by von Békésy that " ... the purpose of the ears is to point the eyes ..." testifies to the central biological importance of sound localization. Among the most important sound localization cues are the acoustic cues created when incident sound waves are diffracted and scattered by the torso, head and pinnae. These cues are captured by the two so-called Head Related Transfer Functions (HRTFs), one for the left ear and one for the right ear. In principle, by filtering sounds with these transfer functions, virtual auditory sources can be made to seem to come form any desired spatial location.

The HRTFs are rather complicated functions of four variables -- frequency, azimuth, elevation and range. Although they have been studied quite intensively for many years, the factors that control the detailed behavior of HRTFs, the source and nature of the person-to-person differences, and the importance of these differences for synthesizing virtual auditory space are still not fully understood. However, this presentation will show how their major features can be explained in terms of simple physically-based models.

Prof. Duda is an Emeritus Professor of Electrical Engineering at San Jose State University. He received his PhD in EE from MIT in 1962, and for 18 years was a member of the technical staff of the Artificial Intelligence Laboratory at SRI International. There he engaged in research on pattern recognition and rule-based systems, co-authoring a well known text book entitled Pattern Classification and Scene Analysis. For his research accomplishments, he was elected a Fellow of the Institute of Electrical Engineers and a Fellow of the American Association for Artificial Intelligence. He is currently a Visiting Professor at the University of California, Davis, where he is engaged in both basic and applied research in the area of sound localization.