Using Compression

If you're like Tom Buck - and most content creators - you're probably turning knobs on your mixer and effects plugins until everything sounds better, but without really understanding exactly what’s going on. And like Tom, you probably know the “secret sauce” for professional audio quality is compression. So audio engineer Marcus dePaula is here to answer Tom’s question about compression by walking us through the 5 main compression settings: threshold, ratio, attack, release, and makeup (or output) gain. 

Download the free MCompressor plugin Marcus demonstrated in the video at https://www.meldaproduction.com/MCompressor

0:00 Introduction

Experienced creators know that to get the best audio quality possible, first and foremost, you need a quiet and acoustically neutral environment, a good microphone close to your mouth and a good preamp and converter with proper gain. But once I have the perfect recording, there is a fourth essential element that can elevate the sound of my voice from moderately clean to mind-blowingly captivating.

And one of my favorite YouTube creators has asked about this exact process. So let's hear his question.

0:37 Tom Buck’s Question

“Hey, Marcus, I have a question about compression, which I will try to compress into a short thing to leave a good impression… I don't know what the heck I'm doing with compression. I know that compression sometimes is that “secret sauce” that helps audio to go from sounding good to sounding pro. But I have always just taken the super scientific approach of turning a bunch of knobs until I think I like the way that something sounds. So if you could give me any advice on where to start when it comes to adding compression to especially spoken word stuff like podcasts and voiceovers, I would really appreciate it. Thanks, Marcus. ”

1:12 What is compression and why use it?

Tom Buck, that is an excellent question and you don't need to feel bad at all about just twisting knobs until it sounds great, because that's what I do too.

But if we can actually understand what twisting each knob does to our signal, ultimately that's going to help us save time every time we record, and it's going to help us get a consistent professional audio quality for all of our content moving forward.

And if you're watching this and you're thinking, “I think I know what compression does, but I'm not sure exactly,” you're definitely not alone. But starting off, we could just look at the name itself. What is the definition of compress? Compress means “to squeeze or press down.” And that's exactly what we're doing with the loudness of our recording.

When I add a compressor into my signal chain, the goal is to help me get a consistent loudness throughout the entire episode, which is especially important when you have more than one person speaking. And just like with everything, I'm keeping my listener in my mind. I don't want them to have to turn the volume up and down throughout the course of my episode. But in addition to that basic loudness benefit, compressors also help me boost the energy of the voice. And to me, that's what gets it up to that professional level, making the listener want to lean in and hear more.

But instead of just talking about these concepts, I want to show you on the screen so you can see what's going on as you listen [see the video above]. So if you don't have your headphones on, go on and grab those really quick because I want to make sure that you can hear these subtle differences.

2:37 Threshold (where)

The threshold’s number value of decibels basically tells the compressor exactly where we want that compressor to start squashing the signal. Now when I record my voice into my MixPre-6 II, I'm trying to hit anywhere between -25 and -6dB on the meters. And those values can correspond to what we're looking at with the threshold. So nothing is going to be squashed if the threshold is set at 0dB. So if I turn it down, then we're going to start to hear the compressor kick in. Because the loudness of my voice is now above that threshold that we set for the compressor to do its thing.

One thing to keep in mind with the threshold is that it's very much dependent on the volume of your recording. So if you recorded things pretty quiet, you'll either have to boost that signal before you insert the compressor or you'll have to adjust the threshold to a lower decibel level so that it starts to compress at that lower volume. And conversely, if you recorded it at a very loud volume, you won't need to turn that threshold down quite as far.

The other thing in this plug in that we need to keep our eye on is the gain reduction. I don't like to see that gain or reduction meter hit too hard. And each plug in is different. But it's usually between -3 and -6dB of gain reduction, showing us how much it's actually squeezing the signal down. And since I'm a “less is more” kind of guy, I don't really want to hear the compressor working that much. So I want to find a level where it's not squashing my voice too hard and it still sounds natural.

4:13 Ratio (how hard)

Part of getting that natural sound is to make sure that the next setting, the ratio isn't set too high. So with the threshold telling where the compressor will start to squash the signal, the ratio is going to tell how much we want it to compress the signal. And as I go up from 2:1, two, 3:1, two, 4:1 to 8:1, and then all the way up to 20:1, you can see that it really starts to hit that compressor harder and harder. So that every dB that my voice goes above that threshold, it's squashing it back down 20 times as hard.

So since this doesn't sound very natural, I'm going to turn it all the way back down to about 2.5:1. Now, keep in mind, every compressor is different, but I do tend to start with it between 2:1 or 3:1 to keep that natural compression sound. For a lot of plugins, these are the two settings that you have and that's it.

5:08 Attack and Release (when)

But to really be able to use the compressor like the pros do, we need these next two settings, which are the attack and the release. These two dials are going to let us choose when we want the compressor to start and then stop compressing the signal. The vast majority of plugins that I've seen out there start off with it set to a low attack (so it's going to attack pretty quickly) and then a slow release (so it continues to compress for a long time).

That's great when your signal is something like a bass or a guitar that has notes that ring out for a long time. But as a creator, we're recording and manipulating our voice. Human speech is super dynamic and each syllable is no longer than 100 milliseconds. That's just 1/10 of a second! So if that release is too long, it's going to compress into the next syllable and it may not compress like we want it to. So if we shorten that release time, that's going to give us what's called a “punchier” sound. It has a little bit more energy to me. But then if we really lengthen that release time, it's going to make it nice and soft and kind of “pillowy” and “squishy.”

So then with the attack, we're addressing the beginning of the signal. And if that attack is really fast and turned all the way down, it's going to also help the sound be smoother because we're squashing down those high energy explosive elements coming out of our lips, as we say our consonants and the beginnings of each word. So then you can hear if we start to turn the attack up, it's going to begin to let some of these consonants at the beginning of each of my syllables through. But if we turn the attack up too high, it's going to sound like it's pumping and we don't want that.

So for the attack and release settings, if you want a smoother compression sound, you'll want to turn the attack down and you want to turn the release up. If you want a punchier, more energetic sound, you want to turn the attack up and you would turn release down.

7:03 Output or Makeup Gain (what)

So even though there's more settings on screen here, I just want to cover one more important one that is actually where the magic happens. And this is on what's called the makeup gain or output gain. So now that we've compressed it like we want it, we want to use this knob to fine-tune where our loudness sits.

7:18 Recap

So to recap, the threshold is where we want the compressor to start to squeeze the signal level down. The ratio is where we tell the compressor how much we wanted to squeeze the signal. And for natural sounding human speech, I like to be between 2:1 and 3:1, depending on the plugin, of course. Then the attack and release let us tell the compressor when we want it to start compressing and then stop compressing. For a punchier sound, do a longer attack and a faster release, and for a smoother sound to a faster attack and longer release. And then the output or make up gain is the setting that allows us to get that volume level back up to a good energy and loudness for our mix.

Keep in mind that each effect plug in can behave differently depending on each of these settings, especially when you're using plugins that emulate the old vintage analog gear, which I personally prefer to use because they can add a really cool color and energy to my mix.

8:10 Ending

Tom Buck, thanks again for helping us kick things off with such a great question. And I want to encourage everybody watching to go check out his YouTube channel and subscribe if you haven't already.

I'm audio engineer, Marcus dePaula and my good friends at Earthworks Audio who make the award winning ETHOS microphone here have partnered with me on Ask the Audio Engineer because we both want to help make exceptional audio quality more accessible for all creators.

If you have a question about studio equipment, software, recording and editing techniques, or if you're running into a problem during your production process that you just can't figure out, record a quick 30-second video with your question and then head on over to asktheaudioengineer.com to submit your question.

Thanks for watching and we'll see you next time.

Follow Tom Buck at himynameistom.com and subscribe to his YouTube channel at youtube.com/@tombuck.

Marcus dePaula is an audio engineer who spent the early 2000’s at Clair Global touring with musicians like Jars of Clay, Amy Grant, Creed, and Michelle Branch, and then the past two decades installing professional recording studios, producing podcasts and live streams for clients, and providing consulting services.