in July 2005
By David E. Marsh
…as electronic architecture?
The phrase “electronic
acoustic enhancement” describes application of loudspeakers,
microphones and electronic signal processing to improve
or enhance the perceived acoustics of a space used
for music performance. This might be a dedicated performance
hall or a conference hall sometimes used for musical performances.
Another common application is to provide a means of varying
the room acoustic characteristics of a venue to accommodate
different types of events. These systems have even been
used to raise the level of crowd noise in sports facilities
to give the home team a greater advantage!
Most people think of room
acoustics in terms of reverberance, so one goal of acoustic
enhancement is to lengthen the reverberation time. Ideally,
the system also should be able to increase the loudness
of a performance, add musical warmth (bass ratio), improve
clarity and control the “aurally perceived”
size of the room (intimacy). Some systems are more effective
than others at simulating early sound reflections
produced by architectural features found in well-designed
concert halls. In this sense, the concept is referred to
occasionally as “electronic architecture.”
Is it really possible for
electronic systems to produce authentic sounding reverberance,
to compensate for missing lateral reflections or to create
an enveloping sound field where the architecture fails to
The quest to simulate room
acoustic effects is older than the spring reverb in my first
guitar amp (that was 40 years ago!). Of course, today’s
high-quality digital reverb units sound a lot better than
my old springs. However, you can’t just “add
reverb” in the house audio system if the goal
is to realistically simulate reverberation. You must
have sound coming at you from all directions. So, you say,
“What about adding reverb and then amplifying the
signal through a gazillion loudspeakers located all over
the room?” To understand why that won’t work,
let’s consider how real reverberation is created.
Sound emanates from an acoustic
source and “bounces” around the space. Each
surface encountered has an effect on the traveling sound
wave, depending on the sound-absorbing properties of the
surface material as well as its size and shape. Some frequencies
are reflected efficiently while others may be diffracted,
scattered or absorbed. Some sound waves travel only a short
distance (e.g., one or two bounces). Others travel
long distances, causing them to arrive much later, weaker
and tonally altered.
The “gazillion loudspeaker
approach” suggested earlier would provide sound from
all directions, but it fails in several important ways to
simulate quality reverberation:
• Unlike the long, drawn-out sequence of sound arrivals
in a naturally reverberant space, the amplified
sound from all loudspeakers would arrive over a relatively
short period (depending on the size of the room).
• Real sound reflections are absolutely unique in
tonal quality in contrast to loudspeakers all reproducing
a signal with the same frequency response.
• Likewise, each reflection varies in level, depending
on the amount of sound absorption encountered in its path.
• And finally, the initial signal is generated by
a single pre-mixed source instead of being “picked
up” at multiple locations in the room and distributed
from each of them in unique ways.
Driving each loudspeaker
with a separate power amplifier channel and applying multiple
signal-processing paths can address the first three issues
mentioned. This provides a way to simulate sound reflections
or reflection sequences from each loudspeaker with unique
temporal, frequency and level characteristics.
Now, we have to address the
issue of picking up the sound in a natural way. Simple:
Install a gazillion microphones all over the room and feed
their outputs into a signal-processing matrix. This offers
flexible routing to a gazillion loudspeaker channels! Now
we’re onto something. But with all these microphones
and loudspeakers working together in the same room, how
do we keep the system from becoming unstable (ringing or
Designing these systems involves
careful layout that ultimately determines the device quantities
and relative distances between microphones and loudspeakers.
As it turns out, it doesn’t really take a gazillion
to get the job done. The trick is in having enough of each
located in enough of the right places to meet the goals
stated at the beginning of this column, and then to achieve
maximum acoustic gain before feedback. System manufacturers
have employed a variety of proprietary methods
to improve acoustic gain; some work better than others.
In practice, far fewer microphones are required than loudspeakers
to attain a significant increase in reverberation time,
but more microphone locations gives better control over
early sound-energy levels. This affects not only the clarity
and intimacy characteristics, but also the enhanced envelopment
provided by early lateral reflections.
Skillful application of this
technology produces remarkable results. But…interested
parties should remember that it is not really a sound system
design. Clients expect results in terms of acoustics.
Therefore, someone experienced in performance-hall acoustics
should be involved, at least in the layout and tuning of
David E. Marsh, a member of Sound
& Communications’ Technical Council, is vice president
of Pelton Marsh Kinsella. He is a fellow of the Acoustical
Society of America and on the board of directors of the
National Council of Acoustical Consultants. Send comments
to him at firstname.lastname@example.org.