My research so far hasn't really resulted in anything beyond "use compression!" - nothing really on HOW to use it.

I use garageband and the compressor has the following settings:

Compressor threshold
Ratio
Attack
Gain

I've figured out what gain does, but I'm really not sure about the rest. I'm not looking for exact instructions, but it would be helpful to get some general ideas at least. Should I aim for - or + on the threshold? Big numbers or low numbers? Different values depending on growls or high pitched screams?

Any tips are welcome.

Have you read this? https://thegaragebandguide.com/compressor

Any specifics then will depend on the vocalist - their singing technique and their mike technique.

Compression is often misunderstood. It does a few things:

- Controls dynamics of a signal (ie, makes soft stuff louder and/or loud stuff quieter)
- adds density/weight/saturation to a signal (compression can add distortion if used a certain way)
- affects the movement or groove of a performance (it can add/shorten sustain of notes, emphasis the beginning or the middle/end of a sound)

For a more simple explanation, you can think of a compressor as a really fast volume knob that follows a set of rules set by you.

The threshold is the level that the original signal needs to hit in order to get a reaction out of the compressor. If your compressor is set to 0db threshold but the signal is reaching -6db, the compress sits there and does nothing. If the signal is peaking at +3db, then the compressor will act on those peaks.

The ratio is the amount of compression applied to the signal passing the threshold. The larger the number, the more compression. 2:1 ratio is a light kiss of compression, whereas 20:1 almost acts as a brickwall.

Attack is how low it takes the compressor to clamp down on your signal once it passed the threshold. Faster attack means that your compressor will hit your signal faster, meaning that the initial impact of the sound will get slammed. Slower attack will let more of the initial transient of the sound, which can make things sound punchy or aggressive.

If your compressor has Release, then you can adjust how long it takes for the compressor to return to the original volume once it dips back below the threshold. Fast release means your signal will leap back to original volume quickly, slower release means it stays clamped down longer.

Gain, of course, is the volume you are adding back to the signal after the compression stage. Compression, by nature, will make stuff quieter. So the gain knob adds volume back.

Hopefully this helps. There are no one-size-fits-all compression settings for vocals, but I tend to go for faster attack and higher ratios. Aggressive vocalists tend to have a lot of spitty mouth noises so fast attack will choke those out. I dial the release to suit the tempo of the song and I dig in with the ratio/threshold until I feel like the vocals become married to the rest of the instruments.

Let me know if you have any questions!

Agree with everything Element_man said above, great summary. I find that contrary to how the end product actually sounds if it's been well mixed, a lot of extreme metal vocals are actually quite soft in terms of sheer volume when you are recording them raw, and so you will naturally end up with quite a few peaks and troughs in the recording that become a lot more pronounced as you gain it up for the track. A compressor really helps to smooth our that volume through the techniques outlined above. I tend to use a compressor fairly early on in the effects chain for extreme vocals as it leaves you with a much more manageable sound to work with afterwards (for EQ and other effects you might want to throw on).

Nice post Element_man! It really will take some experimentation with your vocal style and setup, but that's a solid run down on the tool.


I'm no professional and maybe someone who is could manage it, but yeah! Compressor should be first in the chain, putting say reverb on before compression would create a mess to work with. Get the raw take nice and clean before effecting it.


Indeed. I don't do a lot of growls and tend to do shouty, yelling kinda stuff, and a fast attack helps reduce any plosive sounds of the initial bark. Of course, always best to reduce them on the take so make sure you have a pop filter OP if you don't already!

Do a few takes and then play around with it. As mentioned there are no magic settings, so use Element_man's breakdown and fiddle, see how it sounds better or worse. I do a few runs, slightly tweaking threshold lower and ratio up till I find a sweet spot where it is feeling leveled and even, but not getting to crunchy brickwall.

bonne chance!