Spatial Sound

For my interfaces I use mostly spatial sound. There no big deal about spatial sound as in real life it is always spatial. Which means you can tell, just by listening to the sound, where it is coming from. How does your brain do it? It relies mostly on two cues (out of 7). These two are the Interaural Time Difference (ITD) and Interaural Level Difference (ILD) (the other five are: Reflections from the pinnae and the shoulders, head motion, early echo response/reverberation, vision). ITD simply means that the sound waves arrive at one of the ears earlier than at the other, hence there is a time difference (unless the source is at 0/180 degree azimuth). The brain can compute the direction of the source from this. The ILD is similar: as sound waves are pressure differences in air the head blocks some of them. Let's say the sound source is on your right side - now your head will shield your left ear. This is again detectable by the brain (unless the source is at 0/180 degree azimuth) and helps calculating where the sound source is located.



So when I say I am using spatial sound I mean that I take a monophonic sound source and artificially modify it so that the brain thinks it's coming from a certain position in 3D space. How is that done? By applying Head Related Transfer Functions (HRTFs). Nowadays most sound cards can do that. For my work I use a sound library called OpenAL to do the calculations for me. There is more information about OpenAl here.

There are still some tricky things, like solving the front-back confusion or "cone of confusion". There are some positions which cause similar ITDs and ILDs (same time and same level differences) and the brain can't tell in which of the possible positions the sound source is actually located.

If the position of a sound source has meaning, like it might in an auditory interface, both the front-back confusion and the cone of confusion can get you into trouble.