Last month, we explored the tools and techniques of recording effective narration. This month, we turn our attention to the editing, processing and polishing of narration tracks. Whether you have a full-blown nonlinear editing rig or a simple audio mixer, here are 13 things you can do to insure your recorded narration is as good as it can be.
The snappiest narration won’t accomplish much if your viewers can’t hear it over the din of music, natural sound and effects. Keep these elements low when narration is present, and smoothly bring them back up when the narration ends. You may even need to dip the voice-range (middle) frequencies from your background audio with an equalizer to make room for the narration.
Having heard your narration over and over ad nauseum by the time you’re done editing, remember that your natural tendency will be to mix narration too low. Compensate by pushing the narration up a little higher than is comfortable. Too loud for you is usually just right for the viewer.
Sometimes, breath sounds will seem exaggerated in a narration track. Reducing these loud breaths by 6 to 10dB will make for a more natural sound. Though it’s probably not necessary (and certainly isn’t natural), some videographers eliminate breath sounds completely.
With an audio mixer, simply lower the narration track fader at each objectionable breath. It may take a few practice tries, but you’ll eventually get a smooth track. Folks with nonlinear editors have it easy. Most of them need to do little more than lower the audio track’s amplitude line between spoken phrases.
A dark, boomy narration sound is the norm when using inexpensive microphones and less-than-perfect mike placement. If your narration sounds thick and unclear, don’t despair. Instead, use an equalizer to "cut the mud."
With the simple EQ found on most inexpensive mixers, try turning the bass or low-frequency knob down to about the 9:00 position. If you can select the band’s frequency, try cutting some sound in the 150 to 200Hz range. If you’re using a graphic EQ, make a smooth dip centered around 200Hz. Folks with computer-based editors should be able to accomplish the same thing by adding an EQ "filter" to their audio.
If your narration seems to lack sparkle and clarity, you may need to boost some high frequencies. High frequencies are responsible for the intelligibility and crispness of your audio. They’re also the first to go as you make dubs or pass the signal through low-fidelity equipment.
On a mixer, try turning the high-frequency EQ knob up to about the 3:00 position. If your EQ offers more control, try boosting the range of frequencies around 4kHz. You can also boost all the frequencies about 4kHz and above, but watch out for an objectionable jump in hiss.
At the other extreme, it’s possible for narration to sound thin and tinny. This is common when the mike sits too far from the narrator during recording. The solution is to add some "warmth" and fullness with EQ.
Starting with the simple mixer again, try bringing up the low-frequency or bass control. With an EQ that offers a frequency control (or a graphic EQ), try bringing up frequencies around 100Hz. You may have to boost these lower frequencies a great deal (10dB or more) to hear an effect. As you do, beware a corresponding increase in mechanical rumbling and wind noises.
Don’t be afraid to slice-and-dice your narration, blending parts from different takes into a seamless whole. Accurate editing hardware is a must for this technique, with a nonlinear system being optimum.
The gap between sentences is the logical place to cut, but be careful to not chop a breath in half. The best place to join different takes is usually in the brief period of silence between the breath sound and the first word of a new sentence. Cutting mid-word or mid-breath almost always results in a painfully audible edit.
Crossfading is a quick, smooth change in level from one sound to another. Instead of editing your narration together with straight audio "cuts," try using 5- or 10-frame crossfades to smooth the transitions. An even more complicated form of narration editing, using crossfades pretty much demands a computer-based system.
The key is to use two different audio tracks for your narration. At the end of a sentence on the first track, for example, perform a 5-frame fade to silence. Over the same 5-frame space on the timeline, fade up from silence on your second track. This virtually insures a glitch-free edit with no noticeable change in the amount of background noise.
When editing narration together, be careful to preserve natural pacing. Gaps between sentences or thoughts are a normal part of the spoken word. Cut them out when editing, and the flow of your narration will feel all wrong.
The tendency when editing narration is to cut pauses too short. When you’re focusing all your attention on the narration, a two-second pause can feel like an eternity. When adjusting the pacing, I like to sit back and passively listen to the narration, hearing the cadence, flow and rhythm more than the actual words. Give yourself a little distraction–look out a window, or doodle on a scratchpad. You’ll be better able to make natural edits this way.
The human voice varies in volume as we speak, giving our words emphasis and inflection. With narration, though, quieter words can get lost. An electronic device called a compressor reduces the range between the loudest and softest sounds. The end result is a narration track that seems to float on top of competing sounds without being overwhelmed.
You can insert an electronic compressor pretty much anywhere between the playback deck and the record deck. Computer-based editors can use compression plug-ins or filters to achieve the same results. Learning to hear and set up a compressor takes time. A good starting point is a compression ratio around 3:1, with the threshold set to give you about 6dB of gain reduction at the loudest points.
Can you say, "sssssibilance?" Caused largely by mike selection or placement, sibilance is that loud burst of high-frequency sound that accompanies consonants like "s" and "t." If you reduce high frequencies to compensate, your narration may get too dull.
A special device called a "de-esser" reduces loud sibilance without affecting any other part of the narration. Like the compressor, you can patch it in pretty much anywhere. Computer users can buy de-esser plug-ins to process their digital audio.
Sometimes, you need a narration sound that’s out of the ordinary. This could be a booming echo for the voice of God, or a metallic grating voice for an alien species. A digital effects processor can perform such fanciful feats on your sound, and won’t bankrupt your equipment budget.
In hardware form, an effects processor intercepts your audio signal anywhere between playback and record decks. A "mix" control allows you to set the effect level relative to the dry signal. Computer-based effects plug-ins offer every bit as much fun as their hardware counterparts.
Nonlinear editors offer a key advantage when working with narration: automation of levels. Since the proper narration level in one part of your video may not be the best for another section, you have to adjust volume on the fly. With a computer, you can make the changes once, and the software will remember the correct levels. This feature can save a lot of time.
If you’re trying to match the sound of narration tracks recorded at different times or with different microphones, the automation features of nonlinear editor can be a real help. You can automatically boost, cut or equalize sections as required for maximum consistency. With automation, the only real limits the quality of the audio in your videos are your imagination and ingenuity.
If you want your narration to sound like that of the pros, why not compare it to the pros? Record a few minutes of the evening news or a documentary on tape. Without watching the picture, play back this tape on the same speakers you use for your editing. Quickly switch to your narration, and listen to the differences. Does yours sound thin in comparison? Maybe you need to experiment with equalization. Does your narration seem too loud relative to everything else? Perhaps it’s time to adjust your mix.
You may want to digitize a little clip of great-sounding narration into your nonlinear editing system. Keep it around as a benchmark, and compare your audio to it as you go. In time, you’ll be better able to copy the level, dynamics and tonal qualities of the professional audio. Go to school on the pros!