Audio Techniques for Documentary
How to Capture Good Location Sound
By David Appleby
As with all recording, good audio demands that you keep your subject’s volume as strong as possible (signal) while keeping any competing sounds (noise) to a minimum. What’s different about documentary production is that you will need to accomplish this in situations where you have little or no control over what you’re covering.
In narrative filmmaking, you may be working on a sound stage where noise is almost non-existent. Even on location, you can employ acoustic treatment to limit interference and unwanted reverb. The boom operator and mixer are able to watch the blocking of the actors during rehearsals, so they are able to choose the best position(s) for their microphone(s) as well as know who will be speaking and when.
When the camera and microphone are immersed in the real world in real time, it’s a very different animal and particular skills and equipment are needed to capture good “location sound.”
From the beginning of sound films in the late twenties, through the 1950s, almost all sound for documentaries was created in post-production. There were some experiments setting up sound gear on location but, for the most part, audio recording equipment was too cumbersome to work with in the field. Thus, in most documentaries from the 1930s through the 50s, music and narration were prominent.
Alan and Susan Raymond shooting An American Family in 1971. Photo: Courtesy of Alan and Susan Raymond
By 1960, filmmaker/engineers like Ricky Leacock, D.A. Pennebaker and others had created a “synch rig” that employed a portable reel-to-reel tape recorder that could be synchronized with a 16mm camera. (We refer to this as double system recording.) There was no cable connecting the recorder to the camera, allowing each to maneuver within the space as needed. The recorder could be carried over the shoulder and a “shotgun” microphone allowed for recording relatively clean audio from a distance. This two-person team was much nimbler than the larger crews of the past and they were able to create an intimacy with their subjects that opened up new possibilities for documentary material.
Shotgun microphones have the advantage of being very “directional,” meaning that they pick up mostly from the front while nullifying sound from the sides. So, the sound person can isolate a conversation, for instance, from much of the noise around it even when recording from a distance. That’s good for keeping audio out of the shot but it’s a compromise. The camera is free to frame a wide shot or pan from one thing to another without worrying about getting the mic in the frame, but audio will always suffer when increasing the distance from the source, so there’s a balancing act going on. The subjects are free to move about as they please and that freedom can result in great film – something that may be worth a less than perfect soundtrack.
Photo: Courtesy of Emily Austin Akins
While it may be more distracting than a long shotgun from afar, a shorter shotgun on a boom (technically a fish pole) held above and within a few feet of the subject’s mouth is the choice of a mixer wanting the most natural sounding recording. Handling a boom for a long period of time and knowing where to stand, when to get close and when to back off requires experience, fitness and an intimate knowledge of what the camera is seeing at any given time. In addition to being aware of how the shot is being framed, the boom operator must be careful to avoid casting shadows on the subject or background.
Of course, you may be asking why not just put a small lavaliere mic and a transmitter on your subject, thus eliminating the need for other mics altogether? While this seems like a better option than having a mic either farther away or intruding into the scene, there are reasons many filmmakers avoid them.
These reasons are both philosophical and practical. Some filmmakers have no problem asking a subject to perform everyday tasks, manipulating both time and space in order to capture the images and sounds needed to convey a particular situation. Asking the subject to be “wired up” would be just one more request. Others would never do such a thing, finding it anathema to “capturing reality.” What you do affects what you get.
Documentary legend, Ricky Leacock was always inventing or adapting his equipment so he would be able “to walk into a room shooting.” Nothing would be asked of the subjects except access. Nothing would be staged or repeated. The world of the subject would be kept as natural as possible given the presence of a small two-person crew. This is because he was adamant about how the crew and their equipment affected documentary performance. Connecting a mic, transmitter and antenna to a subject worked against the hands-off relationship he felt so important to protect.
Richard “Ricky” Leacock. Photo: Courtesy of Drew Associates
Ricky once told me that when a film crew making a documentary on the pioneers of Cinema Verité attempted to put a lavaliere under his shirt, he told them to get the hell out of his house!
That said, you may have noticed that all news crews use radio mics now – hand-held and lavalieres – and you may find that miking your main subject with a lav and transmitting that signal to a receiver connected to the mixer or camera allows you the freedom and quality you need. Furthermore, you may find that the subject forgets about the mic attached to their body and performs naturally.
For me, this is a last choice for the following reasons:
(1) Unlike a news crew, I’d want to hide the mic under the subject’s clothing, requiring tape, moleskin, more tape, and a loop to relieve tension. Even with all that, there’s still a chance that the person may create some rustling sound or touch their chest (and mic) at an inopportune moment. Unlike in narrative filmmaking, I don’t like to ask them to repeat something because of bad sound.
(2) The sound is too “close up,” exaggerating the bass frequencies and capturing sound more from the chest than the mouth, all making it sound more like a TV interview and less like natural dialogue.
(3) Radio mics are subject to various types of interference and, eventually, you’ll experience some of these.
(4) When you treat your subject like an actor, they will inevitably become one.
But varied situations call for varied strategies and you should be prepared to use a lav and transmitter when other options aren’t suitable, such as when there are multiple subjects and your audio person can’t reach them all with their boom or shotgun.
Photo: Courtesy of Alan and Susan Raymond
Alan and Susan Raymond were early practitioners of adding wireless mics to the mix when filming verité style for An American Family in 1971:
“Notice that Susan is recording double system sound using a crystal sync Swiss made Nagra 1/4” tape recorder and she is holding a German made Sennheiser 805 shotgun microphone. Also notice she has a backpack on that is holding an early wireless microphone system receiver. The Nagra had three separate mike inputs so she could have three microphones running at the same time though they had to be mixed by Susan into a mono recording at 7 1/2 IPS on a reel-to-reel tape. “ – Alan Raymond
Double System vs. Single System Recording
While news film cameramen from the 50s through the 70s could plug an audio cable into their camera and record sound on a magnetic stripe adhered to one edge of the film, filmmakers have almost always recorded “double system,” the camera recording picture and the recorder (first optical, then magnetic, and now digital) recording sound. The camcorder allowed for audio and picture to be recorded on the same device (“single system”) but the emergence of DSLRs with their generally poor sound quality again saw filmmakers going back to recording audio on separate recorders (better signal to noise ratio, better preamps to boost the mic level signal to line level, and a higher sample rate).
If you’re using a traditional camcorder or a newer digital cine camera, you can record single system, double, or both. Just as it did for documentary filmmakers in the sixties, working with double system allows sound and picture to move independently of each other – no cables to maneuver or trip over. To achieve the same in single system, you’d need to untether yourself from the camera and use a transmitter to send a signal from your sound mixer to a receiver on the camera.
Photo: D. Appleby
A Sennheiser receiver is seen here mounted to a camcorder.
If you choose to shoot double system, remember that you’re going to need to use a slate of some kind to facilitate synching the picture to sound in post. In most documentary situations, the use of the standard clapboard slate is impractical. Besides requiring another crewmember, it’s far too intrusive and distracting to the subjects being filmed.
Slating: Heads or Tails?
Narrative filmmaking techniques allow for a slate at the beginning (the head) of every take. A camera assistant places the slate in the shot, calls out the scene and take numbers, closes the clap stick, then moves out of the shot before someone calls “action.” In documentary filmmaking, the cameraperson might begin filming at any moment. The audio person follows suit if they are not already running (often times, in double system, audio may be recording even when the camera isn’t). Slating, then, needs to occur at the end of the shot. This is called tail slate and it must be done before the camera stops rolling. The cameraman turns to the always ready sound person who whispers a quiet description of what was just shot into the mic and taps it to provide a visual/audio reference point. The camera can then stop recording.
There are other ways to do this. A “bloop box,” for instance, sends a tone to the recorder while flashing a light for the camera. The important thing is that you create a reference point for both audio and picture without disturbing the people or events you’re filming.
Most editing software allows you to synch double system sound to picture by recording with the in-camera mic as well and allowing your editor to synch the two tracks by matching their waveforms. This works fairly well but, even with this method, I’ve gotten more accurate results when the definitive slate sound is present.
Alexandra Pelosi Photo: Courtesy of Bobby George
A directional mic built into or attached to the camera can be useful as a second source, or as a sole source of audio when the cameraperson is working alone. In such a case, you want to keep the camera as close as possible to the source (using a wide-angle lens) in order to isolate it from other sound. There have been some excellent documentaries made using this method. The films of Alexandra Pelosi (Journeys with George, Right America: Feeling Wronged) come to mind. When she’s at a distance from her subjects, the audio track in the final film contains mostly first-person voice over. When her subjects speak, mostly in interviews, the camera is close enough for the on-camera mic to do its job well.
When working alone and concentrating on a single subject, it may be that you’ll need to mic that person with a lav and transmitter to be able to get a usable recording when they’re at a distance or not facing the on-camera microphone.
Always wear headsets! If you’re shooting double system, the audio mixer/boom operator needs to have a headset on at all times. If you’re working single system with audio coming to the camera from a mixer, BOTH the audio person and the cameraperson should have headsets on. Any problem with audio coming out of the mixer, or from a transmitter/cable, will be heard by the cameraman.
Also, remember that most headset connections are allowing you to hear audio as it comes from the microphone. In most cases that works fine but you will not hear audio problems that may be occurring on the recording itself. So: 1) Make sure you listen to playback periodically to check the quality of the recording you’re getting. 2) If you have a device that allows you to monitor the recording rather than the input, use it. (You’re hearing audio played back from the recording device.) It takes a while to get used to the slight delay this creates but it’s worth it to be able to monitor exactly what’s been recorded.
Left: Analog VU Meter
In the days of analog recording you’d want to set your voice levels around O on a VU meter, leaving you a good deal of room above that for louder than anticipated sounds to be recorded without distortion. In such a case, the range of decibels above zero is called headroom.
Left: Digital meter on a Canon DSLR with audio peaking at just under -12 dB
With a digital meter, 0 represents the loudest something can be before the audio falls apart (clipping). So, applying the same principle of leaving room above an average record level for louder speech or sounds, you need to set levels low enough to provide adequate headroom. Taking advantage of digital’s wider dynamic range, most practitioners use -12 dB as a “0 reference” but arguments can be made for allowing more (-20db) or less (-9 dB) headroom. (Audio level controls are always on the manual setting.)
In uncontrolled documentary situations you may want to leave even more headroom to be able to capture higher peaks without distortion.
Another safety procedure is to assign the audio from a single source to two channels on the recorder or camcorder. While sending the same tone to both, you would set channel one to -12 dB and channel two to -15 (or lower). That way, in the case of an unexpectedly loud occurrence, channel one may over modulate but the audio on channel two remains fine.
NOTE: I’m familiar with one recorder, the ZOOM H6 that automatically records duplicate tracks with 12 dB less input gain to “save” your audio when a loud sound distorts your primary tracks.
If your recording device has a limiter on it, you may want to use that as well. Unlike automatic gain control which you should never use, the limiter only kicks in when your volume comes close to clipping. At their best, limiters can work quite well, but at their worst they can negatively affect adjacent levels. If you have this feature, experiment with it enough that you’re confident in its performance.
One More Thing
If you’re working with a two-person crew and you’re in a situation where there’s a lot going on (New Years Eve in Times Square!), the person handling audio needs to be alert to everything happening around them. While shooting, the cameraman tends to concentrate on what’s happening within the frame, so it’s up to audio to be listening and watching carefully, both to get the best audio to accompany what’s being shot AND to be aware of what the camera may be missing.
And don’t undervalue the need for wild sound when the camera isn’t rolling. Just as the camera may shoot MOS (without sound), there are many opportunities for audio to record without camera. Be listening for sound that may offer bits of dialogue or a sense of place and go ahead and record them. These additional on-location sounds can provide information, background ambience and local color, all of which can be invaluable in post-production.
David Appleby is an award-winning documentary filmmaker and professor at the University of Memphis. Since being awarded a Kellogg Fellowship in International Development in 1987, his work has concentrated on community development issues and civil rights. His films have aired nationally on PBS, ABC, A&E, and Starz. http://www.memphis.edu/communication/people/appleby.php
More helpful articles and resources
Documentary | Cameras | Lighting | Audio
Post a Job | Network | Join Discussions in the Filmmakers Forums Online
Subscribe to StudentFilmmakers Magazine
Print Subscription >>
Digital Subscription >>