Fab explains what De-essers are and how to use them on vocals using 6 different plugins and hardware units.
De-essing is one of the biggest voodoo in the mixing world. Sometimes it works... and ssssssometimesssssss it does not. This is why you have so many plug-ins on the market that do just that. They all have different algorithms and work in their own special way.
This is why it helps to have a seasoned engineer show you how they work, show you pros and cons and most importantly point to you what to listen for during this risky exercise.
00:00:07 Good morning children!
Today, were going to talk about
de-essers, lovely little creatures.
00:00:12 Whats a de-esser?
Its a machine that removes S's.
00:00:15 Why do we need it?
Well, S's can be annoying,
especially loud on the vocal
in the middle of your track.
00:00:20 Did we always have de-essers?
No!
Back in the day when people used to record
the vocals this far away from the microphone,
there was no need for a de-esser,
because the capsule didn't freak out.
00:00:30 As the years have gone by, I guess
the recording technique has changed,
and now we record vocals this close,
and capsules freak out.
00:00:37 Ps, Bs, and Ss
Ss are particularly annoying.
00:00:41 Now, you could say: why don't we
just bring them down by hand?
Thats done.
But say your first verse is:
She sells seashells by the seashore.
Or, in French:
Les chaussettes de larchi duchesse
sont-elles sèches, archi sèches,
then you do not want to do that
by hand, you want a machine,
hence the de-esser.
Let me show you.
00:01:00 First, lets listen to our example vocal.
00:01:02 This is Alev Lenz singing
"Dont watch me go" in solo,
she can pull it off.
00:01:07 Listen carefully to the difference of level
between the regular signal,
and the S signal in this take.
00:01:12 Check it out.
00:01:24 Ill play it again.
00:01:25 Now, dont worry
about the noises in the back,
thats her playing Rhodes
at the same time as shes singing,
the clonk clonk in the back,
thats live music, its ok,
but listen to the: "I save myself...
the drowning on your shore."
Ss and Hs are very close in this case.
Check it out.
00:01:50 Alright. Just for references sake,
I could come here, look for my S...
00:01:57 And then, edit that down.
00:02:01 Not exactly my favorite sport.
00:02:05 But it could be useful
for the very difficult situations.
00:02:08 It had to be said.
00:02:10 Now, why dont we just compress it?
Since the S is just louder than the rest,
we can just compress it, right?
Let's try that.
00:02:21 Pay attention to everything,
not just the Ss.
00:02:23 Pay attention to the whole signal.
00:02:34 For me to be able to get the Ss,
I need to compress so much
that everything else gets compressed.
Why is that?
If you look at the waveform,
it makes a lot of sense.
00:02:43 This S here is actually quieter
than the A right behind it,
and way, way, way quieter
than the drowning part here.
00:02:50 And this is what the compressor sees.
00:02:52 The compressor and our ears
are different.
00:02:54 Our ears perceive the Ss much louder,
and much more annoying
than the compressor will ever do.
00:03:00 So what we need to solve the problem
is a compressor that knows
what the frequencies are,
so we can tell it: Hey!
Thats what I want you to compress.
00:03:08 Thats a de-esser.
Let me show you.
00:03:10 Heres a very simple de-esser,
the one that comes free with Pro Tools.
00:03:14 Very simple controls.
00:03:16 Frequency: Where in the spectrum
are you looking for Ss?
Range: How much are you going
to reduce the Ss.
00:03:23 HF Only: That means that instead of
compressing the whole range of the signal,
its just gonna compress the high end,
which supposedly is more transparent.
00:03:33 And then, theres the Listen button.
Crucial.
00:03:36 Why? Because it lets you set up
the de-esser really easily.
00:03:39 What it does is let you hear
what the de-esser hears.
00:03:42 Not the whole signal, just what
you're gonna focus on, where the Ss are.
00:03:46 Because the principle is to say:
compress just that.
Alright? So check it out.
00:03:51 This is with the Frequency all the way
to the top, and Im gonna bring it down.
00:03:57 So, theres not much at 16k.
If I go down to 10k...
00:04:04 I start hearing the Ss.
00:04:06 My guess is that's actually
much lower than that, say 5.
00:04:12 Theres a lot of energy
in this track at 5k.
00:04:15 Maybe the choice of mic,
maybe the singer, I dont know.
00:04:18 Now, lets listen to
what that does on the signal.
00:04:30 Its very efficient.
Too efficient!
Even though I have
the HF Only button in,
which means its not gonna
compress everything,
its just gonna get rid
of the high end of the signal.
00:04:41 Its still too much,
and sounds very processed.
00:04:43 So lets shorten the range.
00:04:47 Same spot, just less action.
00:04:54 Without, it sounds like this.
00:04:59 With.
00:05:03 You noticed that the first S gets caught,
but the second doesnt.
00:05:08 If I were mixing this track,
I probably would decide
that this particular de-esser
is not working for me right now.
00:05:12 Usually it does,
in this case it doesnt.
00:05:15 As a side note,
this de-esser is really an extension
of what a dbx 902 used to function like
and I think it would be interesting to listen
to what a dbx 902 used to function like,
because it was
in every studio in the world.
00:05:27 And I happen to have one!
So I have a dbx 902 inserted as an I/O loop
in Pro Tools on my track,
so you can hear what it sounds like.
Here we go!
Different animal.
Without, it sounds like this.
00:05:56 With.
00:06:07 The 902 is actually better than the Digi
de-esser at catching that second S.
00:06:13 Its got that little darker tone,
I think thats probably
because my 902 needs to be re-capped,
don't tell anyone...
00:06:19 but, not quite what Im looking for.
00:06:21 It still sounds a little artificial.
Lets keep looking for another tool.
00:06:26 There are lot of cool de-essers
around on the market,
they all have different sets of features,
and each and everyone of them
has one different little advantage.
00:06:34 So its cool to have several, so you can
find the one that works for your track.
00:06:38 Here, I have the Massey.
00:06:40 Stevens great plug-ins are really
affordable, they always work awesome,
Customer Service is great,
its worth the investment.
00:06:47 This de-esser is yet
another extension of the 902.
00:06:50 It does a couple of cool things
that I wanna show you.
00:06:53 The first thing is it has this Auto mode,
where whenever you hit the frequency knob,
you go into that
you know, that Listen mode.
00:07:01 So, if I play the track,
it sounds like this.
00:07:10 I think this is about right for us.
00:07:11 Notice that the riding on this box
has no matching whatsoever
with the riding on the other box.
00:07:16 The other box said 5k,
this box says 3.6,
and Im hearing about the same thing.
Why is that?
Probably because
the filter is different.
00:07:23 Dont always believe
whats written on the box.
00:07:25 Let's de-ess this.
00:07:29 Again.
00:07:33 Another cool feature is
if you turn the Listen on,
and hit this button,
you can actually hear the difference,
meaning, what you are removing
from the signal, by itself. Check it out.
00:07:46 Isnt that magical?
So, obviously Im being
a little brutal here.
00:07:53 Ok! That sounds great!
Now, lets listen to the full signal.
00:08:00 It used to be like this.
00:08:06 With.
00:08:10 I still hear the first one
a little bit too much.
00:08:13 So Steven has put
an interesting feature here,
which is a dry/wet signal.
What does that do?
Well, if you consider
the de-essed signal wet,
and the unde-essed signal dry,
you can actually bring back
some of the dry signal
into the wet signal and compensate,
and do a nice balance between de-essed,
unde-essed, everything in phase.
00:08:32 It sounds like this. If I bring
some of the dry, it sounds like this.
00:08:40 Fully de-essed.
00:08:44 These things are subtle, but this is
the difference between a demo and a record.
00:08:48 So if I go a little further
on the dry
Were getting there.
And if I bypass it, as a reminder
Pretty obnoxious! With
It works great, but its not
what I need for this particular track.
00:09:08 I can see it working
for a vast majority of tracks,
but in this case,
I need something a little different
to be able to catch the two different
levels of S's, and that H in the back.
00:09:17 So lets look at something else.
00:09:20 This is the Waves Renaissance De-Esser.
00:09:23 It has one cool trick
that I use when Im stucked like this.
00:09:27 I can go in Side-chain,
just like in the Massey,
its the same as the Listen button
on the Digi, and listen to whats going on.
00:09:33 And, choose my frequency here...
00:09:37 and find out
where my S's are, first step.
00:09:47 That sounds good to me.
00:09:49 Little difference here,
Im using an actual bandpass filter,
meaning that Im listening
to this part of the signal,
as opposed to a high-pass filter
that just goes like this.
00:09:57 It sounds like this.
This is bandpass
This is high-pass
Meaning that the de-esser
will respond to things
that are happening
above my filter frequencies,
whereas with the bandpass,
its just gonna focus on that one spot.
00:10:16 Now, theres two modes here:
WideBand and Split.
00:10:20 Thats the same as Broad,
or HF Only on the 902,
or on the Digi,
those are the two modes.
00:10:27 Same exact thing, different names,
they just wanna confuse us.
00:10:29 Very successful at confusing us!
To use this, I just switch back
to Audio and lower my threshold
until Im happy with the signal.
00:10:36 Not enough...
00:10:38 Not enough...
00:10:40 Let's go crazy!
So, just like the other ones,
its too much for the first one,
and just about right for the second.
00:10:50 But this plug-in has a Range control.
00:10:53 Wait! I can explain!
The Range control is also in dBs.
00:10:57 And what it does is it allows you
to tell the de-esser how far it can go.
00:11:03 You can put a limit
on how much it can de-ess.
00:11:06 No matter how low your threshold is,
and no matter how hard
the de-esser is willing to de-ess,
you can tell it: No, no, no!
No more than 6 dBs, and thats it.
00:11:15 You can do 2 dBs, you can do 4 dBs,
but you can never do more than 6 dBs,
or 5 dBs, or whatever you choose.
00:11:21 Its a great way to make
the de-esser super sensitive
by lowering the threshold,
so that every S,
or anything that resembles an S,
will trip it.
00:11:29 But then, you limit how much
it can trip, and it works like this
Lets try 5,
with the same threshold as before.
00:11:37 Remember, this is without the Range
And this is with a Range of 5 ish
So now, the first S, even though
its much louder than the second S,
and its freaking out the de-esser,
because of the Range, its just getting
a little bit de-essed and it works!
Ill play it again without anything
With the de-esser as is
And as a reminder,
without the Range control
And this is with a Range of 5
Again
Definitely more even
between the two Ss,
and more natural
than what we had before.
00:12:28 This seems to be the right tool
for this problem.
00:12:31 Am I super, super happy with this?
Not bad!
Can I do better? Maybe!
Lets try something else!
Next is the Oxford SuprEsser
from our good friends at Sonnox.
00:12:40 Full disclosure, I had a fairly heavy
hand in helping design this thing,
so it does what I want, so I like it.
00:12:46 This is a slightly different principle,
its broader and does more things,
but you can use it to de-ess.
00:12:51 So the way you use this is
go in Inside mode
and listen to just the part
youre gonna work on.
00:13:06 Maybe somewhere like this.
00:13:08 The cool thing about this one is,
if you noticed, when I play the signal,
theres a red line, vertical,
that shows me where the most energy is,
at that moment, in the signal.
00:13:17 And I bet you,
thats where the Ss are.
00:13:21 You see? I missed it. Im a little
too low, so Im gonna go higher.
00:13:26 A little higher...
00:13:28 See, thats where the S is,
this is where the rest of the energy is.
00:13:33 The D is here. And the S the SH is at 4.1.
00:13:38 So Im able to actually
have a broader band here,
and take care of all that stuff
at once. It's very nice.
00:13:44 Its a little bit like the bandpass
in the Waves,
but more precise and more flexible.
00:13:49 So, the next thing I would do is
turn on the effect,
go back to listen to my full signal,
and lower my threshold.
00:13:58 Lower...
00:14:07 That works nice! Lets focus on
the first two Ss, maybe even lower.
00:14:13 As a reminder, we started here,
with no de-essing
With.
00:14:23 Its super smooth.
00:14:25 You can push it further
by hitting the Access button,
and now you have full control
over everything, attack, damping,
wherever the trigger is in the band,
or actual fully level-related.
00:14:37 It can get really intense.
00:14:38 Theres also a Wet/Dry signal here, so
you can tuck in the dry back under the wet,
like you do on the Massey.
00:14:45 One setting I use is this one,
the Ratio here.
00:14:47 I can actually decide that Im gonna be
a little more brutal on the Ss.
00:14:51 It sounds like this.
00:14:55 So, say Im gonna limit the range,
a little bit like the Waves,
less compression,
but at a lower threshold.
00:15:05 Or a lot of compression
at a higher threshold.
00:15:11 Option A! This just was not very good,
so I switch back to a lower threshold,
and a limited amount of compression.
00:15:22 This is an amazing plug-in, and its not
just for de-essing, because its full range.
00:15:26 So you can actually remove
something of the like of an S,
like some short burst of energy,
automatically, very easily.
00:15:34 I think we should try something else,
just because we can.
00:15:38 I wanted to show you the Eiosis,
because its fairly new,
so not as many people know it,
and I think it has its own merits.
00:15:44 The way it all functions is this way:
Detection, Sibilants and Voiced.
00:15:49 Detection is your key,
the Listen thing.
00:15:52 The Sibilants is the Ss, and then
Voiced is the rest of the signal.
00:15:56 The thing thats different here
is you can actually process EQ
the Sibilants and the Voiced separately.
00:16:03 It can be useful, check it out!
So first, lets listen to the key.
00:16:08 Lower lower
Ok! It understands both,
my first and my second S very well.
00:16:17 And then, once I heard that,
I can say: Oh!
lets just listen to the Ss,
just like I did on the Massey.
00:16:26 Isnt that pretty amazing?
You just hear the Ss.
00:16:33 And then, just the rest of the signal.
00:16:43 The difference here is it didnt care
whatsoever about Drowning, right?
Thats pretty nice! So now...
00:16:51 what can I do with this?
Since I have both signals separated,
I can for example add some brightness
to just the Voiced signal,
but not the S!
So I can retain that aaaahhh thing,
that maybe other de-essers would remove,
depending on what the signal does,
without touching the S.
00:17:10 If I put the two of them back together,
it sounds like this.
00:17:18 And the Drowning is not touched.
00:17:20 Another cool thing here is
I can also process the S,
so if Im listening to just the Ss
Ok I could make the S brighter.
Why would I do that?
Well say if I make the S brighter,
but quieter, it wont feel dull,
it just feels right at the right level,
and at the right color.
00:17:39 Like this, just quiet.
00:17:41 Ok? And the signals together
Maybe even less
Not that much
Without
With.
00:18:00 Its a little more complicated to set up,
and more intense to learn,
but if you set it right,
you can really get the feeling
that you have a special fader
just for Ss,
and thats a good feeling
if you fight with Ss a lot.
00:18:12 So Id like to review
everything weve seen today.
00:18:15 First, the built-in Digi de-esser,
which gives us the basic principle
of how it works.
00:18:39 Then the 902,
one of the original designs.
00:19:02 Then we've seen what Steven Massey does,
with the cool Wet/Dry ratio.
00:19:26 The Waves, with the Range feature,
and the band split.
00:19:48 The Oxford, super precise, razor-sharp,
lots of features, CPU hog.
00:20:13 And then the fairly newcomer Eiosis,
that does tricks that nobody else does.
00:20:37 So which ones the best?
I dont know.
00:20:39 It depends on your track, your Ss...
00:20:42 Do you like a bright UI, a dark UI?
Big knobs, small knobs?
Do you have a fast CPU?
Do you like numbers, no numbers?
Its up to you, really.
00:20:49 This was an overview
of the available features.
00:20:52 Hopefully, now you know
how to deal with de-essers.
00:20:55 Et voilà!
Once logged in, you will be able to read all the transcripts jump around in the video.
Fab Dupont is an award-winning NYC based record producer, mixing/mastering engineer and co-founder of pureMix.net.
Fab has been playing, writing, producing and mixing music both live and in studios all over the world. He's worked in cities like Paris, Boston, Brussels, Stockholm, London and New York just to name a few.
He has his own studio called FLUX Studios in the East Village of New York City.
Fab has received many accolades around the world, including wins at the Victoires de la Musique, South African Music awards, Pan African Music Awards, US independent music awards. He also has received Latin Grammy nominations and has worked on many Latin Grammy and Grammy-nominated albums.
Once again, thank you Fab for this fantastic and useful tutorial. Before, by my own trials with my DAWs I was lost with all the buzzwords ans different jargon.
Vey instructive tutorial. Thank you so much.
alejandro uriarte
2020 Aug 24
Genial Fab gracias por tanta informacion!!!!saludos de Argentina.
Frank Zimmermann
2020 May 13
Thank you so much, Fab! Very helpful! :-)
ludwigmack
2020 Mar 02
I love this guy, what a great energy he has! He goes straight to the point and make his best at being objective at the topics he teaches! Thank you!
In the Mews
2019 Dec 20
Why not just add a dynamic EQ to the channel chain and as they sat cut to the chase ... ?
jon.lo
2019 Jul 20
Interesting. The Sonnox worked best for me, although I think it's fair to say I like a bit more S than you, then again, I know my speakers are kind of forgiving around 8-10 K. I'll certainly be bearing this in min though. You do get a nice smoothness by doing this, it's certainly better than just dipping the EQ. Cheers for the tip!
riko.s
2018 Dec 11
nice
alestorm95@gmail.com
2018 Dec 10
Perfect video thank you!
headstack
2018 Feb 20
Thank you Fab, nice demonstration!
It appears I experienced Subjection Bias while watching the Ren D-Esser.
The display ballistics gave the audio on the first ess, a slightly over compressed sound.
Closed my eyes, and it sounded much better.
This is why I always listen critically with my eyes closed, yet the effect has not been so blatantly obvious.
Fun stuff, the brain, CNS.
Love what you folks are doing, and that you share so kindly with the industry.
Sincerely,
John Chase
jbjohnson937
2017 May 17
I love the R-Deesser! Thanks Fab! Love your videos also. So helpful. Game changing!
DarrylDEwing24
2017 Feb 20
I just got the renaissance desser, so this was very helpful because you listed the plugins used thanks a lot
ezporcelain
2016 Dec 20
Loved all your videos man, incredible stuff! The things you are telling, equipment and the software you use and the way you explain and make your videos. Just Brilliant!
Definitely going to pay for some videos cuz' it's worth it.
One question I have is - does it worth your time to go through all of the deessers to find the right one instead of just automating any 1 of them?
soundspace2001
2016 Dec 09
cool video - just a note: in the EQ video for vocals it says that S-sounds are around 6000-8000 Hz. In this video all S-sounds though are around 4000-5000 Hz. For a beginner a little bit confusing - though you say that frequency ranges are not absolute in all de-essing plug-ins.
soundspace2001
2016 Dec 08
cool
JASONFASHE
2016 Nov 24
Superb!!!
composermikeglaser
2016 Oct 15
Great tutorial Fab, I love that you used a piece of hardware in addition to plugins to tie in the old and new methods of de-essing. The audio example was perfect for this too:)
elmovo
2016 Sep 25
Free de-essing enlightenment, than you very much. Sonnox De-Esser seems to be very versatile... like for using in a backing vocals submix when they peak in some freq (1,9-2,2 khz).
GavinS
2016 Jul 29
Awesome video. I finally get it! Deessers aren't an easy tool to use without a great amount of understanding and listening! Thanks Fab.
rayrothschild@hotmail.ca
2016 Mar 25
very informative-the layers of sound just get more subtle-thanks Ray
izakto
2016 Mar 25
Very good, informative and short tutorial. Fab you're great at explaining how this units work. Great job. Learned a lot. Thanks.
kajman
2016 Jan 06
The king of tutorials :)
kebberj
2015 Nov 12
Hi Fab. What about if you are dealing with both S's and SH's which are at slightly different frequencies? Would you use two narrow band de-essers, one for each problem area? Or perhaps just set one de-esser at the lower SH frequency to treat everything above it?
Thanks very much.
rfelixstudio
2015 Oct 25
Excelente!
Rodrigo Cunha
2015 Mar 02
Bom Demais!!!
VincentDeeper
2014 Dec 22
Excellent tutoriel !! ;-)
daniloco
2014 Nov 24
Exelente tutorial, me encanto el de-esser de Eiosis.
mikeyed
2014 Oct 22
Hi Fab
This was a super video with all these plug ins never saw any of them... Thanks bro.
T&T
Alexander Roca
2014 Sep 05
Excellent video lessons !
AdRi__Forn
2014 Apr 05
Good!!
Peacesound
2013 Sep 27
Hi Fab
Great stuff here. But s il te plait .Veux tu bien arrêter d utiliser le mot "enfant en t adressant aux gens ? C est un peu dérangeant et je ne suis pas le seul a le dire.
Sans rancune.
Neofernando
2013 Aug 29
Thanks for this viedeo, Fernando from Uruguay !!!
beau Jeffries
2013 Jun 27
Great clip. The E2 Eiosis is great too. My fav De Esser. Oxford is great also.
beau Jeffries
2013 Jun 27
Great clip. The E2 Eiosis is great too. My fav De Esser. Oxford is great also.
estephens3
2012 Dec 13
you guys re gonna get my money....this site is brilliant.
joshwoods
2012 Nov 22
Great Video! Fabfilter's new de-esser has been the best I've worked with so far, and exceptionally transparent and easy to dial in.
mrsunnycliff
2012 Jul 07
Great tutorial with super detail, I can't believe Fab & Puremix released this vid for FREE...
shabfish
2012 Jun 23
Could anyone please tel me if it's best to put the 'Deesser' 1st on the channel strip before the EQ or after ?
creaktor
2012 Jun 12
nice to see what different de-essers do. I tend to use the percision De-Esser from UAD. But I think that oxford is well designed with the mind what fits a fast edit and continue fastest to the next step in the mix. More time left :)
Fabulous Fab
2012 May 17
darrendrums: Wouldn't that be wonderful? YOu could but then you loose the proximity effect. For some music it's the best thing, for toehr music it's more difficult. Also most singers have no forward vision about the mix and will freak out because they'll complain that 'it sounds like I am faraway'. I always push singers back as far as they'll stand before they get upset :-)
Fab
darrendrums
2012 May 16
Can't we ask the singer to stand further back ?
besza
2012 Apr 25
thanks Fab!
agent19
2012 Mar 23
Great tutorial, thanks! I'm really interested in Demoing the Eiosis E2 De-esser. Thanks for posting this FAB.
jeandoe
2012 Feb 14
Another VERY useful video! Thanks!
gabrielesancho
2011 Oct 27
Really nice tutorial! Keep on likke this!
Stone
2011 Oct 07
Great tutorial!!! Love it, love it, love it!!!
I have a question for you...are there frequencies that instantly come to mind when you think of solving sibilance problems? So for example when de-essing a male vocal what would be a good starting point for S, Sh, Th, F, etc.?
I never know where to begin and tend to overly rely on the de-esser to compensate for my ignorance! I would love to know where to begin looking without the aid of a De-esser.
PS Thanks for the heads up on the Eiosis De-Esser. It's a fantastic plugin.
111Entertainment
2011 May 30
Excellent tutorial Fab!!!
I will definitely be recommending and posting a link to this site.
"Thanks Again"
"Touch The World"
F.L. Freeman, Audio Engineer/ Mixer
111 Entertainment LLC
islandcode
2011 May 13
great video, excellent presentation and production.