The Big Squeeze: A Beginner's Guide to Audio Compression

For home audio novices, there are few devices that are as confusing as a compressor. Used properly, a compressor/limiter can greatly benefit and improve your recordings, but just like anything else in the studio, too much can make things go wrong very quickly. Here’s a beginners guide to audio compression.

What Is A Compressor?

For the purpose of this article, when we’re talking about compressors and audio compression, we’re talking about devices that limit the dynamic range of recorded audio. Basically you can think of a compressor as an automatic fader – it’s essentially a robot that controls the level (volume) of your recording, based on instructions you’ve provided.

Why Use A Compressor?

There are a number of applications for a compressor, but the most common uses are:

To even out a vocal take or compensate for poor mic or vocal technique.

To even out a bass guitar performance to achieve a more consistent level.

The increase the sustain of an electric guitar.

To reduce the dynamics of a mix, allowing you to increase the overall volume and give a final mix that “in your face” radio sound.

To “fatten” a sound or make an instrument sound more “punchy”.

To limit the signal entering a digital (or tape) recorder, preventing clipping or distortion.

The Controls

Threshold: This setting allows you to tell the compressor when to start working. You’re instructing the compressor to kick in once the audio signal has crossed the threshold that you’ve set. In simple terms, we're talking about volume – once the audio reaches a certain volume (the threshold), the compressor starts doing its job.

Threshold is typically measured in dBs (decibels), with 0 being the loudest before distortion or clipping occurs. Setting the threshold lower applies more compression, for instance setting the threshold at -30dB is going to compress the signal more than setting it at -10dB. If you have a signal entering the compressor at an average of -18dB, with peaks reaching -5dB, and you set the threshold to -10dB you’re only going to be compressing the peaks rather than the bulk of the signal.

Ratio: OK, time for a trip back to algebra class. Remember when you asked your teacher “when am I ever going to need to know the slope of a line?” — well, the time has come. The ratio tells the compressor how much to attenuate (turn down) the signal once it has passed the threshold. If you set a ratio of 2:1, audio exceeding the threshold by 2dB will be reduced by 1dB. If you set a 4:1 ratio, that same audio (which is 2dB above the threshold) will be cut down by 1.5dB. When you set the ratio very high (100:1 or infinity:1), the compressor becomes a limiter, not allowing the signal to go above the threshold at all.

Attack: This setting affects how quickly the compressor acts once input has passed the threshold. The longer you set the attack time, the more transients and dynamics you let through. For instance, you might set a longer attack time to allow the stick hit of a snare drum or the pick attack of a guitar to punch through.

Release: Release time, or decay, lets the compressor know how long to compress the signal before returning the audio to its original, unaffected volume. Longer release times allow the compressor to “hold on” to the signal longer, gradually returning the audio to its original volume. Short release times, on the other hand, will cause the compressor to return the signal to the original volume more quickly, resulting in more of a “pumping” or “breathing” sound. Some compressors offer an auto release setting, which can be quite useful.

Hard/Soft Knee: Often times a compressor will allow you to set a hard or soft knee. This setting determines how the compressor adjusts the level once signal passes the threshold. A soft knee tells the compressor to gradually start working as the signal approaches the threshold, whereas a hard knee setting tells the compressor to compress the signal only when the threshold is crossed. As a result, a soft knee setting tends to sound more natural or “musical” than a hard knee setting.

Gain: Since a compressor’s job is to attenuate the signal, the overall volume is typically lower than the original level. This control allows you to compensate for this by adding output gain. This allows you to do things like leveling out a performance by attenuating the loud parts of the signal and boosting the quieter parts.

Starting Points

Here’s quick reference for places to start with various instruments. These are just starting points, different styles of music require different treatments for compression.

Vocals
Attack: 1 - 10ms, Release: 50 - 200ms, Ratio: 2:1 – 6:1
For more aggressive vocals, try high ratios (4:1 – 8:1) and shorter attack and release times.

Bass guitar:
Attack: 2 -10ms, Release: 300 – 600ms, Ratio: 3:1 – 8:1

Acoustic Guitar:
Attack: 5 -10ms, Release: 100  –  300ms, Ratio: 1.5:1 – 3:1

Electric Guitar:
Attack: 2 - 5ms, Release: 300 – 600ms, Ratio: 2:1 – 6:1

Kick/Snare:
Attack: 1 – 20 ms, Release 20 – 60ms, Ratio: 2:1 – 8 – 1

General/Mix
Attack: ~1ms, 100 – 300ms, Ratio: 2:1 – 5:1

Before running wild and applying compression to everything you’ve recorded, it’s generally a good idea to listen to each track and make some decisions about problems that can be addressed with compression. After you’ve made your assessments, apply compression where needed and be sure not to over do it. Too much compression can suck the life out of an instrument, or worse can boost a lot of undesirable background noise in a track. Be sure to listen to how the effect sounds in the context of a mix; you might have a great sounding, compressed track that sounds right when soloed, but it might not work as well once other instruments are brought in.

There aren’t really any hard and fast rules with compression – the best advice is to trust your ears and experiment. If you can get a handle on compression, your mixes are going to sound a lot better.