Unplugged Sound
Localization 101
Anton Niedersteiner - Acoustic Sound Reinforcement & Audio Engineering
Understanding how humans localize sound is best approached from the survival viewpoint. To begin with let's imagine ourselves outdoors in a forest. Our hearing system has two primary functions: to assist in interaction with our environment and to assist in communicating with each other. Quite literally our survival could depend on how accurately we hear.
Physically our hearing system consists of two nearly coincident transducers (middle ear) separated by an acoustically opaque sphere of about 7" in diameter (head) with crude focusing elements attached (outer ear). It is this rudimentary physical arrangement our sub-conscious uses, processing the countless nerve impulses from each inner ear, continuously comparing the signals to each other and to what was just heard.
Left to right (horizontal) localization is achieved by two basic methods; Time (phase) differences and amplitude differences. Low and mid frequencies have time compared, mid and high frequencies have amplitude compared. The overlap is from about 750 Hz to 1500 Hz wherein both time and amplitude are compared.
This is quite logical considering the wavelengths of the frequencies involved and the shadowing of the head. If a high pitch sound like the snapping of a stick, approaches from an angle to the right for example, then the signal is heard louder in the right ear than the left, the wavelength being sufficiently small enough for the head to provide a good acoustic shadow to the left ear.
Now let's consider a low pitch sound, like footsteps on hollow ground. The wavelengths are longer now so the head is too small to provide any significant shadowing. Since the amplitude at both ears is almost identical, the mind must now determine which ear heard the sound first and by how much.
The outer ear (pinna) focuses very high frequencies (above 6KHz) thereby contributing to changes of timbre. In this way we recognize the sound as coming from in front of or behind us, and to a lesser degree, from above or below our line of hearing (vertical cues).
Further "fine tuning" is provided as we move our head relative to the sound. More comparisons are made and subtle pitch changes (Doppler shift) are perceived. By comparing these signals the mind can quite accurately determine the angle of the incoming sound, to within one or two degrees.
When we retreat into our caves, another psychoacoustic phenomena comes into play. Namely the precedence (sometimes called Haas) effect. It has to do with reverberated sound or echoes. Basically we "fuse together" or assimilate similar sounds that are heard slightly apart but still close together in time, generally within about 30 ms. The original sound and the reverberated sound are perceived as being one sound, belonging to the source which is heard first. The time element varies though, depending on the sound type (attack and decay envelope). Sharp percussive sounds begin to "defuse" with a delay of only 5 ms and steady continuous tones after about 50 ms. After this time two distinct sounds are perceived, one being an echo of the other.
The best way to have the sound appear to come from a certain instrument is to have it actually come from there. Failing that, you could use a few tricks based on the principles discussed here. Stereo mixing boards provide amplitude localization by way of the pan pot. Using a little time delay in conjunction, as the situation warrants, could be the finishing touch.
Visit the Unplugged Sound Website: www.unpluggedsound.com
Back To Unplugged Sound Articles
Website by: Guy Langlois - Webmeister/Publisher Cosmic Debris, 610 Lambert Ave, Nanaimo, BC, V9R 3N8