Comparison of Various Types of Head-Related Transfer Functions for 3-D Sound in the Virtual Environment

摘要

Simulation using virtual reality (VR) is becoming an effective tool for the Army in training soldiers to do their required tasks. In VR, the human operator can interact with a wide variety of computer generated worlds developed from real or imaginary scenarios or both. The training that a soldier receives by simulation is usually cost effective to the Army and in a number of cases is safer for the individual than training in the real environment. Three dimensional (3-D) sound in the virtual environment (VE) provides a more realistic simulation of acoustic environments compared to diotic (mono) or dichotic (stereo) sound presentation. The major benefit of using 3-D sound is that an individual can determine the sound source direction. When sounds that are perceived to have direction and sights that represent virtual objects that produce the sounds are provided through a head mounted display, a person can monitor and identify sources of information from all possible locations. The purpose of this study was to determine if 3-D sound generated by a 3-D sound system could enhance the realism or fidelity of the VE. The main objective of the study was to determine if an individual could distinguish the direction of a sound source within a reasonable degree of accuracy. Three dimensional sound is produced by using a mathematical representation of the filtering characteristics of the pinnae provided through head related transfer functions (HRTFs). The HRTFs can be developed by recording a generated broadband sound using a probe microphone in the ear canal and subsequently dividing the Fourier transform of the recorded sound by that of the generated sound. When digital filtering techniques are used, HRTFs can be applied to sounds through headphones. When an arbitrary sound is filtered with HRTF based filters, the sound should appear to come from specified virtual locations outside the earphones.