How To Use A De-Esser On Vocals (Quick + Simple)

Click Here To Download Your Free Vocal Mixing Cheatsheet
In this video, you’re gonna learn how to use a de-esser on vocals.

But before we dive in, make sure you download my free vocal mixing cheatsheet, which is packed with tips and tricks that will help make your vocals sound more professional. Click the link above or in the description below to download this free cheatsheet now.

Hey, this is Jason from Behind The Speakers, and I have a song here called “Heroes of Hope” by Clean Green Music Machine. And the first thing I want to talk about is why you should de-ess, because a lot of people use de-essers but they’re not really sure what they’re for.

So let’s take a listen to the vocals in verse one, and I want you to notice how loud those s’s are compared to all the other words and phrases in the vocal track. Take a listen.

♪ Seein’ all we are ♪

♪ With superpower vision ♪

♪ A worried world that needs us to keep it level spinnin’ ♪

♪ We’re wise and we’re ready to try ♪

So those s’s just sound like they’re much louder than everything else. And this can happen as a result of mic placement, or boosting top end on an EQ. But a de-esser’s job is to turn down the level of those s’s so they sit evenly with the rest of the words and phrases in the performance.

Now, you might be thinking, well how do I know if the s’s are too loud? I’m not really sure. The truth is, we all have a natural, intuitive sense for what the human voice should sound like. Because we spend so much time every day listening to the sound of the human voice. So whenever you’re unsure, just close your eyes and imagine that singer is right in front of you. And ask yourself, if that singer was right there in front of you, would those s’s sound like they were too loud or would they sound like they were sitting at an appropriate level. And that will always lead you in the right direction.

Now the first thing I wanna mention before we talk about using a plugin de-esser to fix this problem is that one of the easiest ways to solve s’s that are too loud is actually to go through the vocal and just automate them down manually. So whenever you have a vocal where there are only a few problem spots, maybe just a few s’s that are poking out, instead of using a plugin, just go through the vocal track and draw things down with automation.

So what we can do here, if we zoom in, is whenever there’s an s…

♪ Seein’ all we are ♪

So there’s one right at the beginning there that sounds too loud. I’m just gonna draw that down manually with my mouse.

♪ Seein’ all we are ♪

Still a little bit too loud, but sounds much better. So you can actually do this manually, and in many cases if you’re dealing with a vocal where there’s just a couple of spots where the s’s are a problem, this is the most transparent way to do it.

Now obviously, in many circumstances this is gonna take way too much time, especially if you have a vocal where there are just tons of s’s. So in some cases this is not practical. But in some cases it can be a great approach.

So I’m gonna talk about how to use a de-esser now. Which in most cases you’ll find is a much more graceful way to go about fixing this problem. It just takes a lot less time.

And the first question that people often ask is, where do you put the de-esser in the signal chain? Do I put it before the compressor or the EQ? Do I put it after?

And what I like to do is put the de-esser as the last plugin in the line. So after my EQ or compression. And the reason for this is because when we’re de-essing, if we add EQ, for example, after that de-esser and then boost a bunch of top end, oftentimes we’ll find that those s’s get brought out again and we have to apply another layer of de-essing or go back and adjust our de-esser again to fix the problem. So by putting it at the end of the signal chain, we basically can tackle all the s’s just in one spot. And we don’t have to worry about applying different layers of de-essing or going back and fixing the de-esser a million times. It’s just a much cleaner way to go about fixing these types of problems.

So I’m gonna add a de-esser here as the last plugin in the line. And when it comes to choosing a de-essing plugin, now in most cases I’m not a big fan of telling people you need this plugin or that plugin, but when it comes to de-essing I have found that plugin choice does matter. There’s a big difference between de-essers. And some of the stock de-essers, particularly the one in Pro Tools just don’t do a great job.

So I’ve tried lots of different de-essing plugins and I’ve found that this FabFilter Pro-DS is one of the best ones out there, if not the best one that I’ve tried. Now they’re not paying me to endorse this plugin. I’ve just found that this is one of the best tools out there when it comes to de-essing. So if you’re struggling with de-essing, if you feel like you’re not really getting the results that you’re looking for, you may wanna look into investing in a third-party plugin like the Pro-DS.

Now the first thing that I’m gonna do here, once I’ve instantiated this plugin, is set up the sidechain filter. The sidechain filter basically controls what the de-esser is listening to and what it’s using to make decisions about when to pull things down.

Now de-essers don’t listen to the full frequency spectrum of the vocal when they’re making decisions about when to de-ess. They just listen to a very small slice of the frequency spectrum. And they use that slice to make decisions about when to turn things down, when to actually apply the de-essing. So you want to make sure that that slice is the slice of the vocal that contains the s’s. Otherwise the de-esser’s not gonna be listening to the right information, and it’s not gonna be able to make the right de-essing decisions.

So on the FabFilter here, this sidechain filter is down here at the bottom left. On your de-esser it might be in a slightly different spot. But there’s this audition button here that basically allows us to just listen to the sound of the sidechain. Which is basically like listening to what the de-esser is listening to in order to make decisions about when to de-ess.

So I’m gonna solo the vocal. So now we’re just listening to the vocal. And the whole idea here is we wanna set this sidechain filter so that we’re hearing as much of the s’s as possible, but as little of the other parts of the vocal that we don’t want the de-esser to respond to as possible. So it’s this balancing point between we wanna maximize the sound of the s’s and minimize the sound of all the other stuff on the vocal. So let’s take a listen and I’m gonna adjust this.

♪ Seein’ all we are ♪

♪ With superpower vision ♪

♪ A worried world that needs us to keep it level spinnin’ ♪

♪ We’re wise ♪

So I think this is at a pretty good spot right now. I’m hearing a lot of those s’s but I’m not hearing a ton of the good stuff on the vocal that I don’t want the de-esser to start reacting to. But if this filter was turned down much lower, take a listen now.

♪ Seein’ all we are ♪

♪ With superpower vision ♪

So you can hear that we start to hear the character and the tone on the vocal where the s’s are not appearing. And that’s an example of a sidechain filter that’s set incorrectly. So you really wanna make sure to set this appropriately for each vocal that you work with. Because the s’s can appear in different spots depending on the vocal. Male and female vocals often will have s’s that appear in dramatically different places, for example. So you wanna make sure to set this appropriately for every vocal that you’re working with.

So we’ve set that, now I’m going to un-solo the vocal so now we’re listening to the vocal in context with the rest of the mix. Now the next thing that you might ask yourself is, on most de-essers you have a choice between wide band mode and split band mode. These two modes that you can use to de-ess.

Wide band mode basically says, whenever the de-esser hears a problem and turns it down, we want the de-esser to turn down the entire frequency spectrum of the vocal evenly. Almost like you were just grabbing the fader on the vocal and turning it down. Split band de-essing on the other hand, says whenever the de-esser applies de-essing we just want it to take out a chunk of the vocal. Ideally, just where that s is in the frequency spectrum and we want everything else to just kind of remain there.

Now you might think that split band de-essing would be more natural and in some cases it can sound better but in the vast majority of circumstance I’ve found that wide band de-essing is the most transparent way to go about de-essing. In some cases you’ll find with split band de-essing when you start to use split band de-essing and the de-esser just pulls out a chunk of the s but leaves another part of the s in the s’s can start to sound very odd and unnatural. They can start to dull out where you still hear the low end and the mid range of the s but not the top end. It just starts to sound like the character of the s’s changes in a really weird way. Wide band de-essing on the other hand is just gonna turn down the entire s evenly. So it’s a much more natural way to go about de-essing. Almost like if you were going through and just turning down the fader whenever those s’s appear.

Split band de-essing I find more useful when I’m using a de-esser for something that isn’t de-essing vocals. Like you can use a de-esser for example, if you have a electric guitar track where you’re just hearing a lot of that kinda fizzy top end on the top of the track and you can use a de-esser to just turn that stuff down. Or an acoustic guitar where I’m hearing too much of the kinda brightness and the pluck right on the top end. Those are other circumstances in which you might use a de-esser and in those cases I think split band de-essing can be more useful. But when I’m working on vocals, 99 times out of 100, wide band de-essing, I find, is the way to go.

Okay so now that we’ve set that up there are two controls here on the de-esser that basically control how the de-esser reacts. Threshold and range. Threshold basically controls when do we want the de-esser to start reacting to those s’s. How loud do they have to be in order for the de-esser to be triggered. So we wanna set the threshold low enough where we’re seeing that the de-esser is reacting to those s’s.

Range control on the other hand says, what’s the maximum amount that we want that de-esser to be turning down those s’s. So the higher the range control is, the more aggressive the de-esser is gonna react to those problems. Now, in practice I’ve found that it’s best to adjust these controls in tandem and in context with the rest of the mix. So you never wanna be soloing the vocal and making these decisions when you’re just listening to the vocal on it’s own. We really wanna be listening to what that vocal is doing, how it’s sitting with the rest of the tracks in our mix. So that’s what I’m gonna do. I’m gonna listen to the vocal with everything else and adjust the threshold and range controls until I feel like those s’s are sitting in an appropriate spot compared to the rest of the words and phrases in the performance.

♪ Seein’ all we are ♪

♪ With superpower vision ♪

♪ A worried world that needs us to keep it level spinnin’ ♪

♪ We’re wise ♪

♪ Seein’ all we are ♪

♪ With superpower vision ♪

♪ A worried world that needs us to keep it level spinnin’ ♪

♪ We’re wise ♪

So to my ears that sounds much better. So let’s take a listen before and after the de-essing. And again I want you to just listen and zero in on how those s’s are sitting in the vocal track compared to the other words and phrases. This is without de-essing.

♪ Seein’ all we are ♪

♪ With superpower vision ♪

♪ A worried world that needs us to keep it level spinnin’ ♪

♪ Seein’ all we are ♪

♪ With superpower vision ♪

♪ A worried world that needs us to keep it level spinnin’ ♪

So with the de-esser engaged those s’s get turned down and suddenly they sound like they’re sitting evenly with the rest of the words and phrases in the track.

So this is how I would approach de-essing a lead vocal. This is what I find to be the best approach when you’re going for de-essing leads. Now when you’re de-essing background vocals or groups of vocals there are some other things that you need to pay attention to.

So whenever you have a group of background vocals where there’s lots of different tracks that are all kind of singing the same parts, maybe there’s some harmonies, or different parts that are all singing the same line but slightly differently you wanna make sure that you de-ess on the individual track-by-track level, instead of applying a de-esser to the group of background vocal tracks. And the reason for this is because the s’s on all of those individual tracks are gonna appear at slightly different spots. And so if you just apply one de-esser on a group of background vocal tracks, that de-esser is not gonna be able to address those s’s on the individual vocal tracks when they appear. Whenever that de-esser reacts it’s gonna turn down all of the tracks evenly. And so what happens is whenever there’s an s on one background vocal track all of the vocal tracks get turned down. So the de-esser ends up punching a bunch of holes in this big group of background vocal tracks when really you only need to solve the problem on that one individual track where that s appears.

So you can see here I have a group of background vocal tracks but I have a de-esser on each individual track. And I’m using layers of de-essing on each individual group of background vocal tracks instead of just putting it on the bus where I have 20 different background vocals that are all being fed into this group and I’m just applying one layer of de-essing there. This is gonna give you much more transparent, natural sounding results.

The other thing you just wanna make sure when you’re applying de-essing to a stereo track where there are two vocal tracks, maybe there’s a vocal that’s panned to the left and a vocal that’s panned to the right, you wanna make sure that you’re using what’s called multi-mono de-essing. So you can see here in Pro Tools, and your dock might be different there are two choices here when we instantiate a new plugin. We can use multi-mono or we can use multichannel. Now multichannel is gonna link the left and the right sides of the de-esser. So basically whenever the de-esser hears an s, it’s gonna turn down the left and the right sides of this track equally. But if there are two separate performances, there’s one vocal on the left side and then there’s a different take of the vocal on the right side, those s’s on those vocals might appear in slightly different spots. So we actually want the left and the right side of this de-esser to be reacting independently of each other. We don’t want them to be linked. We want the de-esser to make independent decisions about when to turn down the s’s on the left side and when to turn down the s’s on the right side. So whenever you have a situation like this where maybe you have a harmony vocal or a background vocal where there are left right vocals, where there are two different takes on the left and the right side, you wanna make sure that you’re using multi-mono de-essing.

Now I hope you found this helpful and if you wanna dive deeper don’t forget to also download my free vocal mixing cheatsheet which is packed with tips and tricks that will help make your vocals sound more professional. Click the link above or in the description below to download this free cheatsheet now.

And before you go, leave a comment below this video and let me know what’s your go-to plugin chain for mixing vocals? I’d love to hear your response so leave your answer in the comments section below.

Thanks so much for watching and you can check out more mixing videos like this one right here on my YouTube channel or at BehindTheSpeakers.com.

About Jason Moss

Jason is an LA-based mixer and the founder of Behind The Speakers. He's a graduate of New York University's Clive Davis Institute of Recorded Music. His how-to articles have been featured in leading industry publications by Berklee, TuneCore, SonicScoop, The Pro Audio Files, and Disc Makers.