Editor's Note: This is the first of a four part series from Carmen
Borgia on sound for film. The next three parts are an overview on
acquisition, editing/mixing, and outputting of sound for a film.
I first got into sound by way of music, through listening to records.
You know the black shiny things with the grooves in them. I’d buy a new
piece of vinyl at the local department store, hurry home to throw it on
the turntable and utterly bliss out between the speakers or cupping the
headphones over my ears. When side A was finished, I’d flip it over and
play side B. When side B was finished, I’d flip it over to play side A
and so on. I couldn’t listen to a great piece of music hard enough and
I made up tricks to get the sound further into my head: louder, again,
Later in college, the music department had a little recording studio
with some extremely cool devices, among them a multi-track tape machine
and a beautiful analog synthesizer. My dreams were pinned on getting to
those pieces of gear and messing with them. I got my fingers on some of
those knobs; I twisted them and made really cool noises...
These are the two ends of the spectrum of sound work: listening and
making. Listen - take sounds in through our ears for immediate pleasure
and to add to our internal library of things we’ve heard. Make - play
or create or manipulate sounds for us or others to hear. Around and
around it goes.
Sound for film is a special area, because it accompanies images in
time. In film and video, sound people share creative space and budget
with creators of visual elements. When done, it all plays as one single
stream of experience, but for the moment let us consider the unique
aspects of sound.
Under typical circumstances, we hear sound through our two ears, which
sense pressure disturbances in air. Someone speaks and their vocal
chords vibrate like a kazoo (or is it a bassoon?) that in turn vibrates
the air in our mouths and sinuses and then the air all about. This
vibrating air falls upon our ears or a microphone and is translated
into something else. It’s a beautiful, hippy-dippy chain of vibrations
from beginning to end.
The vibrations are three-dimensional; they emanate from their source in
every direction in varying degrees. If the waves encounter surfaces,
they bounce off them and fill up whatever space they inhabit. Air is
not a very efficient medium for sound, and so the waves lose energy as
they travel and bounce around, until they dissipate into nothingness.
The traveling waves are complex, but they interact with physical spaces
in ways that our ears do not fail to recognize. This is why a recording
made in a gym sounds wrong if we play it against an image of the
outdoors. All spaces that contain a sound also color the sound, which
is why one room or environment sounds different from another. There is
no such thing as a sound that does not happen in a space.
Sound in a given space is a dynamic event. When we are listening to
things, we unconsciously move our heads from side to side just a little
to orient ourselves relative to the sound. This is how we can locate
things by sound alone. Our ears triangulate to it.
Ears pass physical sounds as signals into our brains, which interpret
what the ears have heard. The ears and the brain together form a
We can close our eyes and stop seeing something, but we cannot close
our ears and cease to hear. Even when we are asleep, our ears are open,
taking in physical events in the environment. When my partner awakens
me at night, the first question will always be, “Did you hear that?”
When a three dimensional sound hits the business end of a microphone,
the waves jiggle the diaphragm of the mic, and the mic turns them into
an electrical signal. With today’s technology, this must happen before
we can do anything else with sound, such as record it onto tape or get
it into a computer.
The catch is that the “space” that a microphone can accommodate is only
two-dimensional. Though a good microphone can capture a very accurate
representation of the point where the sound hits it, it will
necessarily miss much of the rich information that is occurring in the
rest of the room. Our psychoacoustic systems are dumb enough that they
can work with this arrangement, but a simple test of listening in a
room while someone speaks into a microphone, and then putting on
headphones or listening over speakers will reveal that the experience
of hearing the recorded sound is somehow not the same as the experience
of hearing the original.
Various arrangements have been tried to capture a three-dimensional
sound field, and some are quite good, but none will fool a good
listener. Using more than a single microphone adds dimensionality, but
nothing perfectly simulates the experience of being in a particular
space with sound.
This is a catastrophic situation; we cannot make an accurate recording
of a sound in space! It is, however, not so bad, as it seems. At least
we can get an excellent two-dimensional recording of the sound.
Additionally, there seems to be an acceptance of whatever the current
state of sound reproduction is by the audience of the day. Certainly
all of us have had great experiences while listening to recordings
played back out of a single, small speaker. In terms of film, it may be
enough that the two-dimensional picture on the screen is well served by
the sound of a dimension less than reality.
Someone reading this is saying “Wait a minute, what about 5.1 sound in
theaters?” This setup does in fact add significant dimensionality to
the listening experience, but six speaker channels spread around a room
does not reality make. In a real space, the listener will receive sound
from every direction as the sound bounces about and reflects from every
surface in the room.
Once the microphone has done its work, our sound is transformed into
audio and lives on planet gear. First, you will use a recorder to
capture and store the audio, then an editing system to manipulate and
mix the sound. When the struggle is over, you’ll have to stick the
sound onto a master tape or render it otherwise consumable by your
The recorder shall not color the sound being recorded. It shall not add
much of its own noise and it will work as long as you choose without
somehow losing the audio or breaking down. or having one of the inputs
go bzzzzzz on an entire day of takes.
Audio editing systems these days tend to be all-in-one affairs called
workstations. They use a computer, come in a range of wallet flattening
flavors, and are capable of capturing, transferring, editing, mixing
and outputting audio. Not all workstations are equal. It is likely that
the thing costing a few hundred bucks doesn’t fairly compare with that
which costs twenty grand. As far as I can tell, the most widely used
system is Pro Tools. It ranges in price from five hundred bucks or so
to tens of thousands of dollars. I don’t care to promote Pro Tools
here, as they have their own marketing department, but you are likely
to encounter facilities with those systems. That said there are plenty
of other good systems that are in the hands of competent engineers.
Your workstation should be able to handle your recorded sounds without
altering or degrading them, unless you wish to do so intentionally.
Simply making a copy of digital audio should not, in theory, degrade or
alter the audio data in any way. Processing the sound, such as adding
filters, reverbs, compression and noise reduction should be done so
that the sound plays as naturally as possible, without the addition of
sonic artifacts created in the audio/digital domain. I’m not talking
about the traffic or refrigerator that you recorded on the set, that is
a natural by product of the recording. I’m talking about sonic soot
that is introduced when trying to make those original problems go away.
How do you know when something sounds good or bad in your system? Is it
because the ad in the magazine said it sounds good? Is it because they
used it in Lord of the Rings? No my friend, you know by listening on an
accurate playback system.
When the audio wants to be played back, it must leave its sagging
swivel chair in front of the grimy computer keyboard and get physical
once again. For our ears to pass sound on to our brains, the air first
must surely vibrate, and we must have a thing to shake the molecules of
the air. This would be a speaker.
A speaker is just a disk of paper hung near a magnet, and the box that
holds a speaker is like a little room that colors the sound it creates.
A perceptive individual will note that a speaker does not resemble most
other things that generate sound, such as a drag racer, a pigeon, an
actress or a tuba. All the same, we expect a speaker to accurately
reproduce all of the sounds that an ear should hear. It is a technical
marvel that they sound as good as they do, and yet a speaker does not
sound the same as this reality thing we keep hearing about.
Speakers that reproduce sound in a relatively uncolored manner tend to
be on the pricey side, and to complicate things, the room that they
play back in is effectively a part of the speaker. Thus, a good and
accurate listening setup is a pair of uncolored speakers situated in a
quiet and uncolored room. I won’t tell you here what my favorite
speakers are because there is a range of acceptable models. This may
seem contradictory (or cowardly) but there are legitimate variations in
listener taste that relate to experience and budget. Remember that
recorded sound is an imperfect and evolving technology.
An uncolored speaker will be one that reproduces frequencies evenly
over the range of human hearing. It will not sound as if it is sitting
in a coffee can. You will not check to see if someone threw a blanket
over it. Your brother in law will not hear the speaker and say, “Wow,
that’s some good bass!”
An uncolored room will be one that does not add significant
reverberation or resonances and is isolated from extraneous sounds,
such as computer fans, hard drives and other sonic spoilage like
adjacent editing systems, classic rock and the occasional open window.
If you want to hear accurately, you must be in a sonically neutral
In a poor listening room, many artists resort to headphones, which is a
good compromise. The good thing about headphones is that they eliminate
the sound of the room, which is also the bad thing about headphones.
When audience members finally experience your finished film, they will
probably not be listening over headphones and so they will not have the
same experience you will have had using them. Remember there is no such
thing as a sound that is not in a physical space. To get a repeatable
experience for your end user, you should do your final mix over
speakers in a good room.
Sound occurs three dimensionally in space (four dimensions if you count
Microphones turn sound into audio. In doing so, they squeeze sound in a
three-dimensional space into two dimensions, which is to say you lose
You must handle audio in a digital audio workstation (DAW), where it is
stored, edited, mixed and output.
To hear the finished product, you must play the audio over speakers in
an acoustically neutral room.
Tune in next month for the second of four installments: Acquisition of
Carmen Borgia is the head of audio services for
DuArt Film & Video
in New York City. He oversees a post production sound department that
provides mixing, sound design, restoration, transfer and printmastering.
His department caters to independent projects in all formats from mono
optical up to digital 5.1. Visit Carmen's personal Web site at: