A compressor is essentially a really fast volume control. It looks ahead and, when it sees loud sections coming, turns down the volume, then brings it back up during quieter sections. You control the amount and quality of compression that gets applied to your signal by adjusting attack, release, threshold, ratio and gain. These controls may go by slightly different names in your software, but knowing what each type does is important.

Attack controls how quickly the effect will kick in. For standard narration, a good place to start is around 25 milliseconds (ms). Release sets the time it takes for the compressor to get back to the original volume. Try starting with 100ms for narration.

The threshold and ratio controls are very important, and they also interact with each other. When your signal level rises above the threshold setting you've chosen, the volume will be reduced by the amount of your ratio setting. Set your threshold too low and your audio will begin to sound muddy and overly compressed. Start by setting your ratio to 3:1, then lower the threshold just until you begin to hear a reduction in the louder sections. If this sounds good, you're set. But most likely you'll need to do a bit of tweaking of both threshold and ratio before you find the sweet spot.

For those of us who hated math, a little more about what the ratio setting does is in order. The ratio is simply the gain reduction relation of the input to output. A 3:1 setting means that, for every 3dB of input (above your threshold setting), the output will be 1dB. Higher ratios and lower threshold settings bring up the relative volume of softer elements in your signal, like breaths, sibilance and even noise, which is why it's important to experiment. At its highest setting, infinity:1, the compression ratio is now a limiter, preventing any increase in volume above the threshold level.

Finally, the gain setting controls the overall output of your signal. Depending on your threshold and ratio settings, your average volume will be lower than your original signal. So you will usually need to add back some overall volume. Targeting a final output level of -1dB will give your wavefile some headroom if you later need to make some adjustments.

Caption: Compression Controls (click to enlarge)

Unlike a single compressor that impacts all frequencies, some software allows you to control the compression of separate frequency ranges or "bands." This is an incredibly powerful tool that can really mess up your signal unless you are careful. If you understand the basics of how to apply single-band compression and cannot get the result you need, then this may be your solution. We strongly suggest you experiment first with some of the presets to see if they get you in the ballpark. The VST multiband compressor plug-in that comes with Adobe Audition features real-time color graphics that represent four separate compressor settings, which control as many frequency bands. The Broadcast preset is a good starting point if you want to quickly maximize your narration. Be sure you select the Brickwall Limiter option to prevent signal distortion.

Nearly all of your audio can be improved by using a compressor wisely. The suggested settings we've made will get you started in the right direction. But nothing will replace careful listening, tweaking and listening again. So practice, play and repeat!

Caption: Multiband Compression (click to enlarge)

Contributing editor Brian Peterson is a video production consultant, trainer and lecturer.

To view the tutorial video for The Compression Connection, click here