I learned compression from an experienced touring A1. Having someone help me connect a control to what I was hearing gave the settings a purpose. That's the approach I want to bring into teaching it: hear a change, describe it, then use the controls and meters to understand it.
For a church sound volunteer, the useful question is simple: can I follow the words, and does the voice still sound natural? Compression can help keep louder phrases from jumping out. The skill is learning how much control the voice needs while keeping its expression.
Start with one familiar phrase
I'd begin with a raw recorded vocal that has both quiet and loud sections. Loop the same short phrase. Keeping the performance the same makes it easier to hear what the compressor changes.
Listen with compression bypassed first. Where does the voice jump forward? Where do words get lost? Pay attention to the beginning of each word, held notes, and breaths. Have the learner describe one thing they'd like to improve before touching a setting.
Then bring in the compressor. For this exercise, try a 3:1 ratio and a soft knee, if available. Attack around 10-30 milliseconds and release around 100-250 milliseconds are rough places to begin. Lower the threshold until the louder phrases show a few decibels of gain reduction. Those numbers give us an experiment to listen to. They aren't a finished vocal sound.
A repeatable listening loop
- Listen — Same vocal phrase
- Adjust — One control
- Match — Equal loudness
- Compare — Bypass on and off
Illustration: repeat the same phrase while changing one thing at a time.
Make the comparison fair
Adjust the output or makeup gain so the compressed and bypassed versions sound about equally loud. Otherwise, it's easy to mistake a louder version for a better one. Compare the same phrase several times. Is it easier to follow? Do the quieter details still belong with the louder words?
Change one control at a time
Now shorten the attack and listen to the first consonant of a word. Then lengthen it and compare. Does the voice feel clearer, softer, or more forceful? A faster attack can catch more of the initial edge; a slower attack can let more of it through.
Attack describes how quickly gain reduction builds when the signal calls for more compression. It isn't a timer that waits and then suddenly turns the voice down. The exact response depends on the compressor, so the sound matters more than copying a number.
Try release next. Listen through a held note and into the next phrase. Does the level recover naturally? Are breaths or background sound moving in a distracting way? Give each change enough time to hear it, then compare again.
Keep knee and attack separate
A hand on a fader is a useful picture. Imagine easing that fader down as a singer gets louder. Knee describes how gently the compressor moves into its set ratio across levels near the threshold. Attack describes how quickly its gain reduction responds over time. Soft knee is a gradual transition in level; it doesn't mean slow attack.
Let the meters explain
After the learner describes what changed, look at the gain-reduction meter together. Connect its movement to the words and phrases you just heard. The meter helps explain the sound and gives us another clue about what to try.
Factory presets can be useful starting points, too. I'd use one as a first guess, then listen and adjust it for this singer, this microphone, and this mix.
As I plan the October 24 class, I want brief explanations, plenty of X32 Rack demonstrations from the iPad, and time for people to try the controls themselves. My goal is for someone to leave able to hear a problem, make one deliberate change, and explain why they kept it.
Technical references: FabFilter timing controls; FabFilter dynamics controls; Waves C4 manual, p. 16. Settings above are suggested listening experiments.