How To Compress Vocals - Polished Compression Made Simple

Compression can make or break a vocal recording. Apply too much, and you'll destroy the performance and character. Apply too little, and you could have a hard time producing a track that sounds professional and ready for commercial release.

In the following guide, we'll help you understand precisely what compression actually does so you'll know how to compress vocals like a pro.

When recording vocals, it's important to ensure we can hear every word in the performance, and it's also essential the vocal sits comfortably in the mix. Depending on the style of music and the singer's performance, applying different amounts of compression will almost invariably improve the result.

The skill lies in understanding how compression works and how to apply it appropriately.

What Does a Compressor Do?

Compressors reduce the dynamic range of your signal; they make the louder and quieter parts of the performance closer to each other in level.

Vocals, by nature, are very dynamic and usually benefit from a little, or sometimes a lot of control, depending on the nature of the performance.

Aside from the main task of controlling dynamic range, a compressor changes the transients (in vocals, we call these the consonant sounds). Depending on the type of compressor circuit or plugin used, it can also colour the sound in a characterful way.

Too Much Compression Could Cause Problems

When applying compression, it's important not to overdo it. Too much compression will accentuate breath sounds or increase sibilance.

You also don't want to squash the life out of the performance. Depending on the style of music, the amount of stylistically appropriate compression will vary greatly, but invariably we want to retain a degree of human expression.

Use Gain Automation Before Applying Compression

In many instances, I recommend applying gain automation to the recording in your DAW before applying any compression.

This initial step is a favourite trick by pro engineers to even out a performance and reduce any potential breath issues that may cause the compressor to work harder.

To benefit the compressor, this automation must occur at the beginning of your signal chain.

Channel gain is not the same as output volume. So adjusting the volume fader will achieve nothing if we want to change what our compressor receives.

Evening out the gain of a performance using automation will make setting the compressor threshold much easier. If the level coming in is very erratic, we could struggle to find an appropriate threshold to suit the louder and the quieter parts of the recording.

As compression shapes the tone of a sound to some degree (particularly the transients), having the compressor work very hard on the loud chorus parts (for example) and only subtly (if at all) on other sections could lead to a dramatic variation in vocal tone.

For the purpose of example, we're using Logic Pro, which has a gain plugin you can place on the insert of any channel. As pictured below, we can then use automation the adjust the input gain before going out to a compressor.

Apply Deductive EQ Before Compression

Reducing any unwanted or excessive frequencies before applying compression is another way to improve the result. Excessive low-frequencies, in particular, carry a lot of energy that will make your compressor work harder than might otherwise be desired.

How To Use a Compressor for Vocals

Once you're happy with the signal you plan on sending to a compressor, it's important to understand the basic parameters.

There are six main controls on a compressor: Threshold, Ratio Attack, Release, Knee, & Make-Up Gain.

Threshold:

Threshold sets the point at which dynamic processing is applied. It sets the level (in dB) above which the compressor acts on the incoming signal.

If your input signal is nice and even, then setting the threshold to catch peaks or apply a broad but gentle compression should be fairly simple.

Ratio:

Ratio changes how much compression is applied when the signal exceeds the threshold.

Expressed in decibels (dB), a ratio of 2:1 means for every 2 dB the signal goes over the threshold, we reduce it by 1 dB.

Therefore, a signal exceeding the threshold by 2 dB will be reduced down to 1 dB above the threshold. At the same setting of 2:1, a signal exceeding the threshold by 8 dB will be attenuated down to 4 dB above the threshold.

Higher ratios result in more attenuation and less dynamic range. Ratio settings of 3:1 - 4:1 are considered by most to be moderate compression.

5:1 - 7:1 results in medium compression and anything above 8:1 gets into the realm of strong compression.

Anything above 20:1 starts to get toward what we consider as limiting rather than compression.

Attack:

The attack setting affects how quickly the compression is applied, or in essence, how long does it take for the signal to be fully compressed.

Release:

The release setting affects how long it takes for the compressor to stop compression.

Knee:

This determines the sharpness of compression. Specifically, it refers to how the compressor transitions between the non-compressed and compressed states. A soft knee is subtle, while a hard knee is more aggressive and up-front.

Gain:

Output gain or 'make up gain' simply allows us to make up for the signal attenuation with additional volume after compression is applied.

Compression Starting Point

It's impossible to give a one-size-fits all approach to setting up a compressor. There are many different approaches depending on the style of music and performance.

Below is a starting point from which you can start to hear the effects of compression and make adjustments based on the style and performance.

Start by applying a fast attack of around 10ms and a medium release time (somewhere between 50 & 100ms).

Next, apply a medium ratio and start to bring down the threshold until you achieve around -10dB of gain reduction. As this point, you'll be able to clearly hear the effect compression is having on your signal.

With this in place, listen carefully to the consonant sounds and adjust the attack until you think the vocal has enough forward edge or delivery to suit the track.

Speeding up the attack will make the vocal sound more aggressive or energetic, and lengthening this setting will produce a more laid back or natural sound.

When you're happy with the tonal quality, you can dial back the threshold a little to produce a more moderate reduction and then use make-up gain to compensate for the amount of level reduction.

Consider Using Two Compressors to Get an 'Up-Front' Vocal Sound

A great trick for achieving an ultra-clear, consistent, and up-front vocal sound is to spread the load over two compressors.

Set the first compressor up to catch peaks and allow the second plugin to apply a more broad but subtle compression. The result sounds really up-front in the mix and often sounds more natural than a single compressor working much harder.

A popular combination for this approach is the classic UA 1176 followed by a Teletronix LA2A - both plugins are based on classic outboard hardware and feature in our best plugins for vocals guide.

Conclusion - How To Use Compression On Vocals

When considering how to compress vocals, it's important to think about what you're trying to achieve in the mix. Resist the desire to jump straight in with presets and controls, and take the time to understand how each control will shape the sound.

Use your ears and listen to how the vocal sounds and sits within the context of your mix.

For example: Is there a huge difference in volume between the verse and chorus? If so, then you definitely want to consider gain automation before applying compression.

Does the vocal sound overly aggressive for the style of music? If this is the case, consider a shorter attack to try and tame some of those consonants. The attack and release settings are really where the tonal shaping takes place, so take plenty of time to adjust these settings until you achieve the best sound for your recording.

Lastly, if you notice some words are still unclear in the mix after compression, consider applying small amounts of level automation to really fine-tune the intelligibility of your vocal.