Learn advanced compression techniques in order to process and compress your vocal track.
Fab discusses how to compress a vocal track without ruining it.
Chapters: -Concept and workflow setup -Standard Avid Compressor -Threshold vs Ratio -Attack vs release -Low level passages handling -Automation vs compression
The attached zip file contains the stereo instrumental and raw vocal tracks of the song.
Please download it and import the files in the DAW you use to make music (any software will do really!) This way, you can improve your mixing skills and transparently compress this vocal track using your own compressor.
Parts of this site and some files are only accessible to pureMix Pro Members or available to purchase. Please see below our membership plans or add this video to your shopping cart.
00:00:07 Good morning children!
Today, we're going to talk about
transparent vocal compression.
00:00:12 Most records do just fine
and actually sound better
with a more compressed tone,
because that's the style.
00:00:18 Older records - different style -
actually benefit
from a more virginal,
or undoctored tone.
00:00:23 I know people who'd benefit
from that too...
00:00:26 Today, I'm gonna take the lead vocal
from a jazz track
and compress it to the level we need
to get the tone we want
but try and keep it as natural
as possible.
00:00:35 Here we go.
00:00:36 This is a track I mixed.
00:00:38 Singer's name is Sara Lieb,
it's her record
producer is Matt Pierson, they're both
amazing, please check them out!
It's a perfect example
for what we're doing
because it's a wide open sounding record,
it's supposed to sound undoctored.
00:00:49 I'm gonna play you the first verse
so you get a vibe for what we're going for,
and then, we'll go into the processing.
00:00:55 Here we go!
Cool!
So, I just want to point something out,
completely unrelated...
00:01:44 I love this...
00:01:46 They used a Bass Rhodes to double
the upright.
00:01:49 A Bass Rhodes is the bottom half
of a Rhodes, in a little case.
00:01:52 And I think it sounds great!
So I'll play it for you.
00:01:59 It's awesome!
And then, you add the bass.
00:02:06 It's really cool!
But it has nothing to do with
what we're talking about today.
00:02:10 So let's go back to that.
00:02:12 The first thing we should do is listen
to the vocal raw, uncompressed.
00:02:15 Listen to the base level, just
how loud it is on average
and then to the peaks, those words
that stick out.
00:02:21 Do they bother you?
How much do they bother you?
Check it out.
00:02:49 She's obviously an amazing singer
but still, she's got emotions,
not a machine
so she leans in, leans out,
there's some louder and quieter parts.
00:03:01 So I propose we try and tame them
with a compressor.
00:03:04 To keep it fair
to all parties involved
we shall use the standard Digi compressor,
which is actually quite great.
00:03:10 How do you think about this?
Well... I usually pick...
00:03:15 a standard ratio, say 2:1, whatever
worked for me last time, if I remember
and then I'll start
lowering the threshold
to figure out where I want it
to happen.
00:03:24 Now, in the spirit of
keeping things natural
I want to compress
as little as possible.
00:03:30 For example, those first two phrases
"When I want you, in my arms"
I don't want to touch that,
those are great.
00:03:36 So that's my base level.
00:03:37 I'm gonna set my threshold
so that the compressor...
00:03:40 only takes care of everything
that's louder than my base level.
00:03:44 Fair? Here we go!
I don't have a number. I can't guess
this stuff, I gotta try.
00:03:50 I'll try here...
00:03:58 So far, so good.
00:04:03 That's a little light. So maybe
if I lower... say here... 21.
00:04:07 Let's check this out. I'm going
straight to the louder parts.
00:04:13 Pretty good.
00:04:26 So, I like that!
It's actually compressing
when I want it to compress...
00:04:30 but not enough.
00:04:32 Some of the peaks are still
too loud for me.
00:04:34 So, then I'm gonna...
00:04:36 augment the ratio.
00:04:38 Meaning...
00:04:39 I'm happy as to which part
of the signal is compressed
but I want more compression.
00:04:44 To do so, I'm gonna raise my ratio.
Say, let's pick a number...
00:04:47 Say 10:1.
00:04:48 Here we go.
00:04:49 Loud parts only.
00:04:54 Wow!
I don't like that!
I think it's too much, right?
So... Oh, you may like that!
That's where mixing branches out.
00:05:12 You have your taste,
I have my taste.
00:05:14 I can't account for that.
But...
00:05:16 I'd say that's too much, in the spirit
of what we're trying to do.
00:05:21 So, let's go back down.
00:05:23 How's 3, and change?
That's quite nice, because it keeps
the vocal right here
it doesn't do this anymore.
00:05:43 And it doesn't sound too squished,
I like it!
So that was threshold and ratio.
00:05:49 Obviously the most important ones.
00:05:52 Let's talk about attack and release.
00:05:54 In this particular compressor,
I have control over attack and release.
00:05:57 Some compressors,
like for example an LA-2A...
00:06:00 no control over attack and release
which is interesting to think about.
00:06:03 Maybe people love the LA-2A on vocals
because the attack and the release
are actually, magically, set just right
for vocals. What's to think about?
In this case, I have choice,
which is great.
00:06:14 We're gonna be able to compare
different settings.
00:06:16 In this particular case right now,
I have 2ms on the attack
and 200ms on the release.
00:06:21 I get a lot of tracks from producers
where the attack is all the way down
to the minimum possible.
00:06:26 Why is that?
But that makes sense!
Instinctively, they're lowering
the attack all the way down
which means that the compressor
is compressing everything.
00:06:36 What does that do to the sound?
Let's check it out!
So this is where we're at right now.
00:06:51 And this is the same,
but with the attack all the way down.
00:06:55 Listen...
00:06:56 to the transients.
Listen to the "hhaa" part of the vocal.
00:07:00 Listen to the presence.
00:07:02 Where is my vocal now,
with a short attack?
Just so you remember, this is...
00:07:18 how it sounds with an open attack.
00:07:37 Now, if you listen carefully...
00:07:39 it's subtle! But you'll notice
that the vocal goes from...
00:07:42 here with an open attack
to here, further back,
with a closed attack
because we don't have
that presence thing
because all the transients
have been eaten by the compressor.
00:07:51 Now, I kinda dug...
00:07:53 some of what the short attack did
for the overall gain control.
00:07:59 So what I'm gonna try
is to get that back
but still have the presence.
00:08:04 To do so, I leave my attack open
and I'm gonna lower my threshold.
00:08:07 Which means that the compressor
is gonna start working
earlier in the dynamic range...
00:08:13 the same amount, with an open attack.
00:08:15 And it sounds like this,
check it out!
I think it's a little too much.
Maybe if I lower the ratio a bit?
Check this out.
00:08:51 I think that's a good balance.
Now, it's tricky!
You have the threshold, so how early into
the dynamic range do we start acting?
You have the ratio, which is,
how much of it, above the threshold
do we nuke the level?
And then, there's the attack, it's like...
00:09:05 how much of the presence
do we keep in?
And then, to make things
further simple, there is the release.
00:09:11 Can it be important?
Oh yes! Check it out.
00:09:15 Right now, I have a release
of 200ms.
00:09:17 We just heard it 4 times,
I won't play it again.
00:09:19 What happens
if I shorten the release?
Listen to...
00:09:24 the steadiness of the vocal.
Has it changed?
Or, is it as centered as it was before?
Listen to the space in between words,
or the end of words. Check it out.
00:09:50 So now it's starting to dance again.
00:09:52 It's starting to, on the "Dream" word,
kind of peek through
and it's not as centered and solid,
and kind of like...
00:10:00 corseted, if you will, in place.
00:10:02 I'll play that again, and then
right away, I'll play a longer release.
00:10:06 Short release first.
00:10:43 Isn't it crazy to think that
by changing that one setting
which by the way is the setting
that most people overlook all the time,
the whole sound of the whole vocal
is completely changed?
There are ways to think about these things.
I'm trying to walk through them
and we do more in the dedicated
compression videos, but...
00:11:01 every setting has
a certain musical impact on the tone.
00:11:06 Being able to think about it
in musical terms helps a lot...
00:11:09 rather than looking at numbers.
00:11:12 This is the way I think about things
when I compress a vocal.
00:11:15 Do I have the presence?
If I lack presence,
I push the attack open.
00:11:21 Do I have the steadiness?
If I want more steadiness, maybe
I'll open the release a little bit
so it's a little more together.
00:11:28 Kept together.
00:11:30 Are my quiet phrases left alone,
or are they squished?
Threshold...
00:11:35 Are my loud phrases taken care of
enough or not?
Ratio.
When you think of it that way...
00:11:40 it's not so daunting
to think that every single parameter
has an effect
on the other parameters.
00:11:45 It becomes a more natural and more
instictive way to setup the system.
00:11:50 Now, I'd like to listen
to these settings
within the song, because
ultimately, that's what it's about.
00:11:56 But before we listen to the vocal
within the track
I'm gonna apply some make-up gain
on the compression.
00:12:01 Why is that?
Well, if all you're doing is just
tame a couple of peaks a little bit
and you're happy with
where your base track is in the song
you don't need make-up gain really.
You're just taming some peaks, right?
Your base level is cool,
you're taming the peaks.
00:12:15 In this particular case, I'm kind of
squashing the vocal quite a bit.
00:12:19 Even though you don't hear it much,
I'm squashing it.
00:12:22 I'm probably gonna have to give some
make-up gain, so that's it's loud enough
on the verses, and stays just
where I need it to be on the peaks.
00:12:29 And I saw on the gain reduction meter
that I lost 6 to 8dBs of gain
on the peaks.
00:12:35 So I'm gonna compensate by about 6dBs
and see how that goes.
00:12:38 Here's the song.
00:13:04 So that's kind of perfect!
It sounds kind of like a record.
00:13:07 There's still one thing
that bothers me a little bit.
00:13:10 If you notice here,
after this loud phrase
there's a very quiet phrase.
Check it out.
00:13:23 So, "All I have to do"
is getting lost.
00:13:27 Here, there's two possibilities.
00:13:29 I can be lazy.
00:13:31 And I could take the threshold down,
so that the peaks are reduced further
so that I can bring
the whole track up
and then, this won't no longer
be quiet.
00:13:40 That's the road most traveled,
I find.
00:13:43 Although I am very lazy,
I tend to not do that.
00:13:47 Because it's gonna squish the whole
vocal, just because of one quiet phrase.
00:13:52 I don't like that!
In this case, I would compensate
by using automation
take just this one phrase
and raise it a couple of dBs
and see how it sounds.
00:14:10 It actually could stand
another .5.
00:14:12 Yes, you can hear .5dBs!
Ok, .8!
So that works!
It is perfectly fine, and sometimes
highly regarded
to use a combination of compression
and automation
to set your vocal levels.
00:14:36 It also works, not only
for quiet parts, but for loud parts.
00:14:39 Say for example...
00:14:41 you have a peak that goes through
your compressor settings.
00:14:44 You have the right settings for the average
thing, but there's one annoying peak.
00:14:47 You could compress more, but then
the whole track gets affected.
00:14:52 You could use a two-step
compression system
which we'll discuss
in a different video.
00:14:56 Or, you could just go in
and automate that peak.
00:15:00 It takes more time,
but it's more natural sounding.
00:15:03 Since that's what we're trying to do,
that may be the way to go.
00:15:05 In this case, I'm happy with the way
the whole verse sounds
especially with that
little quiet part raised.
00:15:11 So that's one way
to think about compression
in the context of
a more natural sounding vocal.
00:15:19 Again, there's not one way
to compress vocals
there is, how do you want it to sound,
and how do you get to that spot
and there's a variety of settings.
This is one of them.
00:15:28 Et voilà!
Once logged in, you will be able to read all the transcripts jump around in the video.
Fab Dupont is a Grammy 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 been nominated for Grammys 6 times, including two Latin Grammys and has received many other accolades around the world, including Victoires de la Musique, South African Music awards, Pan African Music Awards and US independent music awards.
Parts of this site and some files are only accessible to pureMix Pro Members or available to purchase. Please see below our membership plans or add this video to your shopping cart.
Thank U for this, Fab! Wish there are more tutorials about Jazz recording and mixing.
soundspace2001
2016 Dec 08
Very well explained!
Manley
2016 Jun 19
hi Fab
what's your views on tracking vocals with room sound v tracking very dead sounding vocals pro's & cons
rmoreno
2014 Oct 13
Muy bueno el video me abre la mente a mas posibilidades en como ver las cosas.
Fabulous Fab
2014 Jul 04
@soundgroove
: not if you are looking for a very contained sound on a very dynamic recording.
soundgroove
2014 Jul 03
Thanks for this! But isn't 2ms Attack still to fast?
bolupona
2014 May 28
Thanks for that breakdown. Truly appreciated!
AdRi__Forn
2014 Mar 18
Todo muy claro.¡Genial!
bolupona
2013 Nov 28
Love how you explained the different components of a compressor and what each stands for. And also love the fact you used the compressor that came with the DAW. That way we don't tend to get carried away about the gear but focus more on the concept behind the gear.
Thanks.
pirouss
2013 Sep 25
Salut Fab!
J'écris en français, ça sera plus simple pour nous! ;)
Tu parles du LA-2A qui est parfait pour la voix, j'en conviens!
Ne penses-tu pas que c'est du au fait qu'à l'origine, c'était un compresseur d'antenne (pour la diffusion radio)? ( source: http://fr.audiofanzine.com/processeur-dynamique/universal-audio/teletronix-la-2a-classic-leveler-collection/editorial/tests/the-next-generation.html)
Comme en radio, on s'occupe beaucoup, pour ne pas dire essentiellement, de la voix, il y aurait une logique à celà...
Qu'en penses-tu?
Pirouss - technicien RadioFrance
Becky Jo Benson
2012 Nov 07
Fab!
You are so good about details and your videos are very easy to learn quickly because of the way that you talk about
how it boils down to the FEELING of the recording that a person wants to get across. THANK YOU so much for being
there for each of us out here who are learning how to listen! Hahahaha, and I really like your video on 'How To Listen'.
Actually each of your videos have helped me so much to be aware of what to listen for. In one of your videos you said
that if we can't hear the difference, listen to it over and over again until we can---good call! Becky Jo Benson
digilando
2012 Oct 28
Please ... do not saw the Rhodes !
ALEXISCEDENO
2012 Jul 14
Server not found: rtmpt://stream.puremix.net:80/puremix WHATS HAPPEN WITH THE VIDEO?
jam92189
2012 May 28
I bought the how to hear and eq after those i had to subscribe I love these videos
Larryfileccia
2012 Feb 12
I got the Eq Vocals Video, Loved it. and you mentioned Compression. That brought me here. I though I had the Comp on Vocals down. I was doing good but this video has lots of tidbits and ways to think about it and do it that really helps Thanks!!!
android
2012 Feb 10
When you say, "We do more in the dedicated compression videos", which videos are you alluding to? Have they been released or were you referring to to-be released videos like the 2 stage compression one? Great video by the way. I learned a lot.
Adrian Tello
2011 Dec 11
Really nice tutorial. I’m waiting for a 2 stage compression tutorial too.
dcherouny
2011 Nov 11
Nice sound!! Very clean.
SimL
2011 Nov 01
nice tutorial, thanks
and what about the knee? how this feature can be useful?
