AudioTorch: using a smartphone as directional microphone in virtual audio spaces

Abstract

Mobile audio augmented reality systems can be used in a series of applications, e.g., as a navigational aid for visually impaired or as audio guide in museums. The implementation of such systems usually relies on head orientation data, requiring additional hardware in form of a digital compass in the headphones. As an alternative we propose AudioTorch, a system that turns a smartphone into a virtual directional microphone. This metaphor, where users move the device to detect virtual sound sources, allows quick orientation and easy discrimination between proximate sources, even with simplified rendering algorithms. We compare the navigation performance of head orientation measurement to AudioTorch. A lab study with 18 users showed the rate of correctly recognized sources to be significantly higher with AudioTorch than with head-tracking, while task completion times did not differ significantly. The presence in the virtual environment received similar ratings for both conditions.