Compression on vocals is a necessity for many reasons. Without the right compressor settings for vocals dialed in, you might not even notice a difference.
A compressor is a tool like any other. When you know how to use it properly, you’ll get the best results. With that in mind, let’s talk about the perfect compressor settings for vocals to dial in to get that perfect sound every time.
Make sure you check out my vocal EQ guide (which includes a cheat sheet) to cut out all the stuff you don’t want before you hit the compressor. That way we’re only compressing the good stuff. Now, back to the vocal compression!
The Best Compressor for Vocals
Before we get into compressor settings for vocals, make sure you’re using the right type of compressor on them. I recently did an overview on the best compressor for vocals and explained why a FET compressor like the 1176 or an optical compressor work so well on vocals.
And in case you were wondering how many compressors to use on vocals, you can’t use too many. You should certainly be using more than one.
This distributes the compression load between multiple compressors, yielding a more transparent compression. I typically use Arturia’s 1176 modeled compressor as the first in my signal chain on vocals. This does most of the heavy lifting for me, and it features the normal settings featured on most compressors. With that in mind, let’s talk about the best compressor settings for vocals.
Best Compressor Settings for Vocals
Here is my vocal compression cheat sheet that I always use when I compress a new vocal.
Let’s cover each of these compressor settings for vocals in depth.
Threshold/Input
Some compressors label this as threshold, others as input. Whatever name it goes by, this is the level which the audio must reach before the compressor begins reducing the signal.
The ideal threshold is the vaguest question to answer as this will vary from track to track. Instead, set it so that you achieve a gain reduction of 5-10dB on the loudest peaks. Note that gain reduction is achieved through a combination of the threshold/input AND ratio, so let’s get into that.
Ratio
One of the most important settings for vocal compression is setting a ratio at the right spot. The ratio is essentially to what degree the compressor is going to pull down those peaks of your vocals. Check out my quick guide on compressor ratio explained for more information, but essentially a higher number will reduce the signal more.
A high ratio affecting a lot of input will significantly reduce the dynamics in your vocal. This is the difference between the quietest and loudest parts.
When compressing vocals, I generally aim for somewhere between a ratio of 4:1 and 8:1. If you’re using a compressor with a dial, try starting around a 5:1 ratio.
This means that for every 5dB which exceed the input/threshold we set, it will be compressed to 1dB.
Again, the real takeaway is that a higher number means any signal which is compressed gets compressed harder. This takes the dynamics out of your vocal and gives you a more constant level.
As the human voice is naturally a dynamic instrument, more compression sounds less natural, but it keeps that vocal in the front of the mix. No words or syllables are lost when a signal is heavily compressed.
Typically on FET compressors you have the 1176 style ratio presets.
So rather than having a ratio dial like on most stock compressors which come with DAWs, you have locked in 4, 8, 12, and 20 ratio options.
I typically go with either the 4 or the 8 ratio. If the vocal is more dynamic or the genre is such where there’s a greater emphasis on the vocals (like pop), I’ll go for the 8. If I’m compressing a gentler signal or a vocal in a sparser mix and I really want transparent (undetectable) compression, I’ll go with 4:1.
Again, the main goal with the first compressor I’m using is to achieve 5-10dB of gain reduction on the loudest peaks.
Attack
The attack setting on a compressor dictates how long after it crosses the threshold that it begins to be compressed.
This is measured in milliseconds, even as quickly as a fraction of a millisecond which realistically means instant compression.
The whole point of using a compressor is because we want the audio compressed, so why not set the attack to instant?
Setting the attack to be too fast means that all we hear is the compressed audio. In other words, the compressor will clamp down before the initial “punch” of the audio is felt. Those transients help create a punchier individual audio track and a mix overall.
This is why I like to set the attack so that we get to hear the initial punch right before that compressor clamps down on the signal. It’s more of a feel than a hear sensation, and it’s an important aspect of creating a lively engaging mix.
So, to answer the question what’s the best attack on vocal compression, the answer is typically relatively fast.
If you have the option to set a specific time, try starting with 10ms and move it left and right to hear the difference.
The signal will sound flat when you set that attack as fast as it will go. Slow it down a bit and you hear a more natural sound with more of the dynamics retained.
Similar to the ratio, a faster attack also keeps that vocal right in your face. With that in mind, you might aim for a fastest or instant attack in certain genres like pop when you want the vocal to remain consistent through the track.
Let’s get back to the compressor I’m using in this entire example.
Note that the attack is another area where 1176 compressors (which again to remind you are great on vocals) throw you a curveball.
In going against the norm of every other compressor, a higher number on an 1176 means a faster compressor. The attack speed is represented with numbers between 1 and 7 with 7 being the fastest. So if you’re compressed a track and you want it right up in your face, set it at or near 7.
The Arturia gratifyingly even makes it clear right on its interface that higher numbers denote faster attacks.
Release
If attack determines how quickly the compressor starts working, the release dictates how quickly the signal is released from compression after dropping below the threshold.
The ideal release on vocal compression is more clear cut: you’ll typically want a fast release. Try 50ms on a stock DAW compressor. Once again on an 1176, higher numbers means faster release times. Set it at or near 7.
Why?
Because a slower release compresses more signal than necessary/doesn’t let go when it should. This results in an awkward, uneven compression in most cases, at least in the context of what you want.
Having a relatively quick release also preserves the kind of dynamics you want post compression, keeping an energy to the vocal track.
Output Gain
The output gain is simply the level of audio after the compressor has done its thing. You’ll typically want to have the same perceived volume coming out of the compressor as you had going in without the compression. As such, you’ll need to turn up the output dial to put some volume back into the track.
One thing to mention is to make sure that whenever you’re experimenting with a setting like the attack or ratio that you are aware of the volume.
For instance, slowing the attack means more of the uncompressed signal comes through, resulting in a louder volume. This can trick your ears into thinking that a slower attack is better, because louder is always perceived as better.
If you’re comparing settings like that and the compressor doesn’t have an automatic makeup gain, create a second instance of the compressor with the different setting and adjust the output gain manually to match the first instance.
Then you can split test between the two settings by way of the two different compressors and have a realistic idea of which actually sounds better without the louder of the two winning.
After your initial compressor, I recommend following it up with a second compressor to smooth out the remaining signal.
As any awkward jutting out peaks will likely have been addressed with the first compressor, this compressor is just for a bit of glue.
Optical compressors like the Waves CLA-2A are perfect for this role.
You don’t even need to worry about settings outside of the threshold – attack, release, and ratio are all automatic.
Set the peak reduction to achieve 1-2dB of gain reduction on average and at most 3dB of gain reduction on any remaining areas which poke out.
After all of these compressor settings for vocals are set, your vocal should be sitting in a good spot. Follow it up with a bit of automation in case any words aren’t poking through enough in the context of the rest of the mix and you should be good to go.
If you want to add another compressor for a touch more glue, drop it at the end of the chain after any additional processing.
Vocal Compressor Settings Tips
- The best vocal compressor settings yield a controlled, thicker, and more energetic, yet natural sounding vocal.
- Set your compressor threshold or input when dealing with vocals to achieve 5-10dB of gain reduction at loudest peaks.
- Remember to set your output gain manually to match any changes particularly when split testing settings if compressor doesn’t feature auto makeup gain.
- Set a ratio of 5:1 on most compressors. On an 1176 the preset “4:1” or “8:1” settings each work well on vocals. Aim for the “8” for a more dynamic vocal which needs more control.
- Setting a relatively fast attack of 10ms (or around “5” on 1176) is a sweet spot for maintaining punch while keeping a thick, energized vocal.
- Setting a relatively fast release of 50ms (or around “5” on 1176) avoids overcompression and keeps a natural signal post compression.
- Remember to use more than one compressor on vocals. An 1176 is a great compressor first in the chain, followed up by an optical compressor.
Pingback: Serial Compression - The Ultimate Guide - Music Guy Mixing