Manual

Chapter 25 Mixing 559
Binaural Hearing—Binaural Recording—Binaural Panning
An important part of mixing audio signals is the placement of individual sound
sources at different spatial positions. Most common recording and mixing techniques
simply send a given signal (at different levels) to the available loudspeakers—two for
stereo, or more for quadraphonic or surround setups—in order to create a virtual
sound stage.
This approach is somewhat flawed, however, as human beings are able to locate
sound sources at different positions with just two ears. Virtually all spatial information
for all sounds is included in the two signals arriving at the two ear-drums. From these
signals, human beings can determine characteristics such as; inter-aural time and
level differences and—based on the listening experience—information about the
spatial origin of the sounds being heard. Are they coming from in front, behind, to
the left or right or above or below? This ability to perceive where a sound originated
from is referred to as binaural hearing.
In theory, the spatial positioning of any sonic experience can be reproduced during
playback, so no special techniques need to be employed during recording. There is,
however, one drawback to this approach: every person has differently shaped ears,
and different body and head proportions. All of these influence the way sound
signals arrive at the ear-drum—not to mention aspects such as hearing loss,
experiences of the sounds being heard, and so on. Given these physical differences,
each person listening to the same sound source (while standing/sitting at the same
position) will hear slightly different binaural signals.
Therefore, perfect reproduction would only be possible if you could make a recording
with tiny microphones placed inside your auditory canals. As this is impractical,
binaural hearing has been emulated in sound laboratories by using mannequin heads
with built-in microphones. This approach has led to average person binaural
recordings, that are more or less compatible with the way most people hear.
Playback of binaural recordings is best suited to headphones, ideally combined with
signal conditioning (processing) that ensures the signals are accurately reproduced.
Given a specialized listening environment, it is also possible to reproduce these
signals with loudspeakers, utilizing a process known as cross-talk cancellation.
As you are unlikely to have the technology required to make binaural recordings,
your best chance of simulating binaural signals is by processing the sound signal on
playback. This is known as the HRTF (Head-Related Transfer Function), which
approximates the change that a signal undergoes on its way from the source to the
ear-drum.