ImmersiveSound: Exploring Spatial Audio with DSP Tools

Motivation: A lot of basic audio systems use single-channel mono sound, or “fake” a stereo effect by doubling the same channel for a two-channel system. However, this gives the sound a very basic “centered” sound quality. However, humans experience sound through two separate ears on opposite sides of our heads, and we can generally distinguish the relative positioning and direction of sounds based on the differences in certain qualities of the sound we hear, such as its phase. There are also various natural filtering effects caused by the structure of our ears, which can be referred to as a Head-Related Transfer Function (HRTF), as well as effects such as pinna gain (caused by the pinna, the large outer-facing component of the human ear) and more. In this project we wish to understand how spatial sound presents itself in data, and how to reproduce it by converting any input-sound to have our desired positional quality.