Digital filtering for 3D binaural sound

The traditional binaural technique consists in recording sounds using miniature microphones placed at the entrance to the ear canals of a replica of the human head (dummy head microphone). The idea is that subsequent listening to these signals through headphones will reproduce the original auditory experience, evoking the sensation of a source located at a point in the space surrounding the listener. A method of synthesising the binaural technique is desirable in order to avoid its restriction to high quality musical recordings: architectural acoustics, for instance, would benefit from a three-dimensional reproduction system, allowing the designer to simulate the acoustical properties of a performance space and judge them by means of listening tests before its construction. Communication systems can also be improved by the inclusion of spatial auditory information. In teleconferencing, for instance, the parties involved could be effortlessly identified from the spatial location of their voices. This paper describes the implementation of a digital synthetic binaural reproduction system at a minimum computational cost.