What Is Sound Design?
If you hear it in a movie, and it’s not
music or dialog, it’s sound design. It’s
the sounds of a street with honking
cars, people yammering, and raging
ambulances—the sounds of life
moviegoers expect to hear in a scene.
For this article, we will design a lakeside
scene, using the famous painting
by Georges-Pierre Seurat, “Seine
Grande Jatte.” I am using this still
image because it’s readily available to
view online, making it an easy reference
for the sound cues to follow. In
the real world, this would be a film
snippet of a few seconds or more.
Analyzing the Scene
When you view the painting, you’ll
notice a number of elements that
require the sounds of motion: the
water, the sailboat, and the oarsman.
We also have a number of things in
the scene that don’t require primary
motion-oriented sounds, but do need
to provide background sounds and
ambience, such as the wind through
the trees, the flag in the distance, and
overall environmental sounds.
Obtaining Audio Assets
Let’s start with the ambience, as our
scene is going to need some “environmental
tone” and some “water tone.” I
could go out on location with a mic and
a field recorder to capture these
elements, but it would take too long,
and all the trouble wouldn’t buy me that
much. So I go online to one of the
many free sound effects libraries
(such as www.pacdv.com/sounds/
ambience sounds.html ) and pick some
options to use as sonic foundations.
Water, Water . . .
First up are the water sounds. I chose
an outdoor audio clip with a quiet
ambience because the lake in the
painting is still. To build on the source
sound—and to create the appropriate
sense of oars in the water and a sailboat
cutting across the surface—I fill
up my bathtub with water, and set up a
Shure SM57 on an articulated boom
stand. I use a SM57 because of its narrow
harmonic range and tight pattern.
An articulated mic stand is also an
important tool, because it allows me to
put the mic in different positions
around the tub and room to capture all
the water sounds I need.
I decide to design the sound for the
sailboat first. This is a very leisurely
boat, so I drop the mic as close as possible
to the surface of the water, and I
move my fingers slowly through the
water. First one finger, then two, and so
on until I get a variety of “bow through
the water” sounds. I track this with a
light compression setting of a 2:1 ratio
at a –10dB threshold—just enough to
make my “hand waves” sound even
Next up is the oarsman. Oars slapping
the water are more aggressive
sounds with herky-jerky natures, so I
move the mic up about six inches,
and place it off-axis from the tub’s
waterline. Then, I make a fist and pull
it through the water to make some
more violent waves. For this sound, I
use a heavy compression setting of 15:1
and –30dB, and I also crank up the mic
preamp to capture all of the nuances
of the water motion.
Lastly, I move the mic about four
inches from the edge of the tub and
about eight inches above the waterline.
This position is to capture the sound of
water splashing against the sides of
both boats. Now, I splash the side of
the water in different ways to represent
the different sizes of the boats in
Air & Oars
Now I need the sound of wind for the
sails and flag, as well as the sound of
the oars, the man rowing, and the
breeze through the trees. To simulate
the air sounds, I place the mic about
five-feet high at the opposite end of
the bathroom, and I wave a towel in
front of it. I experiment with different
compression settings in order to have
a “wind library” to choose from when I
compile my sounds for the scene. I also
“flap” the towel at different levels of
intensity to collect some flag-blowingin-
the-wind effects. To emulate the
breeze through the trees, I gather
some small branches from outside, and
I wave them gently at the mic from
about two arm-lengths away.
For the oar sounds, I take one of the
rings of my shower curtains off, and
grab my toilet plunger. I place the
plunger through the ring of the curtain
holder, and I set it on the side of the
tub. Then, I hold the ring while I “row
the plunger” through the water in
rhythmic sequences—just as an actual
rower would do. As I do this, my last
effect is to breathe hard, because that
oarsman is exerting himself.
I need two different reverbs running:
a long and wide one (a large hall with
a very fast attack and a long tail),
and a medium one (a medium room
with a fast attack and a smaller tail).
These two reverbs will simulate physical
distances for the scene. The trick
to this is making sure the end result
doesn’t sound like two reverbs of different
sizes and decay times, but as a
seamless outdoor environment that
matches the idyllic scene in the
All of my bus sends from the original
sound-source track will be set to
pre fader so that I can have independent
control over how much wet signal
is used. I send the shorter reverb
through a bus to the larger reverb. This
allows me to have an element that has
its own small reverb envelope (as most
things do), while also moving through
a larger reverb space (or environment).
The water/ambience sample I downloaded
from the Web will run the
length of my “scene” (about 30 seconds).
I send the file through a bus to
the larger reverb only, putting the
reverb sound at about 50 percent and
the dry source sample at about 25 percent.
This simulates a body of water
that is close and far at the same time—
perfect for “distance” cues for the sailboat
(far) and the rowboat (close).
The flag exists mostly in the big reverb.
I use one of the busier towel recordings,
and I fade it in and out of the
reverb over the course of the scene.
This will emulate the sound of wind
gusts hitting the flag.
My sailboat sound—which is comprised
of a combination of the different
“fingers moving through the water”
and “water splashing again tub”
recordings—will be sent to both
reverbs simultaneously. The plan here
is to slowly pan the source sounds and
the reverbs from left to right to match
the “movement” of the sailboat.
My rowboat sound is composed of the
rhythmic oar recordings (which are
the loudest component), the watermovement
recordings (my fist), and
the grunts (which are mixed lower than
the oar-in-water sounds, but louder
than the water recordings, and appear
every three or four strokes). Obviously,
the water sounds, while the lowest signals
in the package, will always follow
the stroke of an oar.
Finally, I will take those breeze-inthe-
trees recordings, send them
slightly to both reverbs, and leave
much of the source signal in the mix.
And there you have it—a “scene” that
has been completely sound designed
in the home!