Vocal Compression: 6 Tips to Know

It’s hard to imagine a vocal track without compression.

Indeed, by nature, vocal recordings exhibit significant dynamic variations that often need to be compressed.

For example, to better fit into a mix when it comes to singing — or to maximize intelligibility when it comes to voice-over.

In other words, compression is one of the unavoidable steps when mixing a lead vocal.

However, this is not an easy task: while equalizing a voice is already complicated, compressing it can prove even more complex.

Because let’s not forget: if you compress a vocal track poorly, you can literally ruin your mix. This can lead to poorly controlled dynamic variations in your track, or the sensation of a voice lacking naturalness.

In other words, even if you master the settings of compression, the exercise of compressing a voice may not be straightforward.

Hence this article, in which I have compiled 6 tips that seem absolutely essential for properly mixing vocal tracks.

Note: if you are not yet comfortable with compression, know that I offer a course on Projet Home Studio that will explain everything you need to know on the subject and thus greatly improve your mixes.

👉 If you’re interested, click here to discover the course.

1. Use automation to better compress vocals

“To compress correctly, you must not rely solely on compressors.”

This may seem paradoxical, but if you understand what I’m about to explain, it will necessarily help you integrate your vocal tracks into your mixes. Especially if your productions have a somewhat modern style.

Often, on the scale of a song, the dynamic variations of your vocal recordings are significant.

For example, one can imagine that the singer gives maximum energy during the chorus, while the introduction of the song consists of almost whispered lyrics.

The result is that you will get a waveform somewhat like this, with large variations:

Now, let’s imagine that you add a compressor to level the voice a bit and ensure that all syllables come through correctly during the chorus. You set your plugin to achieve maybe 5 dB of gain reduction… but only during the chorus.

What happens now during our introduction with whispered lyrics?

Well, the level is much lower, and probably below the threshold you set.

Result: the compressor does not work on the intro. Your signal is not affected at that moment.

So, no compression.

So, no leveling of the lyrics during the intro.

So what to do?

This is where automation comes into play.

Automation is this feature present in most DAWs that allows you to vary a setting throughout the song.

If there are many ways to use automation, it is particularly possible to automate the level of a track via a gain plugin.

The procedure is simple and consists simply of:

1. Add a gain plugin;
2. Create an automation curve;
3. Raise the gain level of the track at the weak passages (sometimes, a few decibels are enough!).

And there you go, now your compressor reacts correctly, even on that famous introduction with whispered lyrics… 🙂

Tip: some plugins allow you to perform the manipulation in a relatively automated way. This is particularly the case with the Vocal Rider from the famous brand Waves (-10% with this link and the code that appears).

2. Choose the attack of your compressor wisely

For me, when we talk about compressing a voice, the attack is the most important parameter.

The attack is simply the speed at which the compressor reaches the ratio you have set.

For vocal mixing, this is absolutely critical.

Let’s imagine, for example, that you set an ultra-fast attack, say below 1 ms (as I’ve seen recommended in online articles…).

What will happen?

Simply put, all the dynamics, all the “punch” contained in the first 20-30 milliseconds of the syllables — the transients, if you will — will be shaved off by the compressor.

The result is that you will feel like your vocal track lacks life, or that the sound is very flat and lacks energy.

That’s normal: the attack was set incorrectly.

Of course, I must moderate my statements: if the ratio is very low, the negative effect will be much less pronounced.

That said, it clearly highlights the fact that choosing your attack correctly is crucial for compressing a vocal track:

• If you want a sufficiently sharp, punchy sound, choose a rather slow attack to let the transients through (> 10ms);
• If, on the contrary, you really want to target the transients to control dynamic ranges and peaks, then reduce your attack. But do so cautiously, to avoid distorting your recording.

Note that a fast attack is not always a negative thing: in any recording, there are peaks that stand out annoyingly and are likely to cause your mix to clip. If that’s the case, a good technique is to use a compressor with a very fast attack and release to clip the signal quite surgically.

3. Equalize the voice before compressing it

A mistake I see very often among beginner home studio users is compressing without having equalized the signal beforehand.

If, of course, your recording is perfect, with an excellent microphone and great acoustics, then this type of practice probably doesn’t pose too much of a problem.

Unfortunately, in a home studio, having an absolutely perfect recording is rare: there are often small frequency issues, resonances, or other background noises. It could be a specific resonance from the microphone, or the noise from your computer coming through between syllables.

If you compress the signal directly, what happens?

Well, you will undoubtedly bring out these sound issues.

It makes sense, really: by compressing the louder parts of the recording, you will bring them closer (in terms of sound level) to the quieter parts.

In other words, to take the example I mentioned earlier, the difference in level between the sound of the voice and the background noise from your computer will be smaller after compression than before compression.

To avoid such situations, I recommend following a three-step method instead:

1. First, equalize your track subtractively, that is, by removing the problems you hear;
2. Then, compress the voice as you wish;
3. And only after that, equalize the track additively, to boost, for example, certain frequencies you like.

This way, you will avoid bringing out the problems of the recording and your compressor will react more naturally.

► To learn more about the subject, check out my article on the order of EQ and compression plugins

4. Use multiple compressors

How many compressors do you use on your vocal tracks?

One only? 🙂

Well, maybe you need several.

I say “maybe,” since the method of compressing a voice depends primarily on the recording and what you want to achieve in terms of mixing.

However, if you are just starting out and are only using one plugin thinking “OK, I will only use one compressor because I’m not very sure how to set it up…” — then it’s possible that this choice may hinder you more than it helps.

Let me explain.

When compressing a voice, we often aim to achieve two things:

1. control the peaks of the signal;
2. and level the sound to avoid too much dynamic range, which would harm the intelligibility of the lyrics.

If you use only one compressor, you may struggle to manage all of this at once because the threshold (threshold), ratio, and attack/release settings will have difficulty adapting to both situations.

The solution is to use two compressors in succession.

Often, “the pros” will lean towards a strategy like the following:

1. first, a 1176 type compressor, very responsive, to capture the peaks with very fast attack and release settings;
2. then a LA-2A type compressor, slower, to level the signal.

Note that with the 1176 placed first, it will facilitate the work of the second compressor by handling the peaks: the LA-2A will therefore react less to this type of dynamic variation.

Of course, the famous analog compressors like the 1176 and LA-2A are just one combination among others: you can very well use plugins of other types, or even use the basic compressor of your DAW twice (I particularly think of the one in Ableton Live, which is really good).

5. Compress the voice just enough

Behind this somewhat naive paragraph title lies a reality that we often tend to forget: there is not just one way to compress vocal tracks.

There are no magical settings, and there is no level of “Gain Reduction” that must absolutely be reached.

However, before adding a compressor to one of your tracks, it is important that you ask yourself what you want to achieve.

This is true all the time during mixing, but it is even more so for vocals.

Specifically, it is important to decide on the style and attitude of the vocal track you want to achieve. Depending on the musical universe you are in, you may have completely different targets:

• for jazz, you will probably use very little compression because the dynamics of the performance are the priority;
• for pop, you may choose more pronounced settings to make the voice stand out amidst the arrangements;
• and for modern pop, RnB, or hard rock, you will likely lean towards more extreme settings to have a voice that cuts through the mix without issue.

Having an idea from the start of the sound you want to achieve will help you choose your compressor and set it to the right level.

In other words, if you want a dynamic performance because there is space in the mix, simply smooth the voice with a few 2 or 3 dB of gain reduction. If, on the contrary, you want something aggressive, for a very “front-facing” effect — then push your compressor to 10 dB of GR or more!

6. Compress while listening to the mix

If there is one mistake I made very often when I started learning to mix, it was compressing my vocal tracks in solo.

That is, without listening to the rest of the mix.

Are you also guilty? 🙂

Well, in some cases, it can be useful. To clearly hear the impact on the frequencies of a compressor, for example.

But most of the time, listening to a track in solo while compressing is a major mistake that can only harm the mix.

To understand why, we need to ask the question of the purpose of voice compression. Why do we compress a voice?

There can be several answers, but generally it is to homogenize its level (by reducing the dynamic range) so that all syllables, all words are clearly audible and understandable in a mix.

Of course, this goes hand in hand with good equalization of the vocal track.

If you compress your track in solo, you simply have no reference to judge the amount of compression you want to apply or the attack/release settings you will adjust.

So even if in solo your compression sounds good (which is completely possible), it is completely decoupled from the reality of the mix. As soon as you bring in the other tracks, you then realize that your track does not fit properly in the mix.

Moral: whenever possible, compress your vocal tracks while listening to the entire mix, rather than in solo.

To go further

There you go, with all these tips, you should be able to approach voice compression a little more calmly.

But don’t forget that this is generally not the only effect used on vocal tracks… check out my guide on mixing vocals for more info! 🙂