Would like to know peoples thoughts and real world experiences with deessing and compression. I have heard/read opposing views. Some say if you heavily compress a vocal your trying to de-ess you will bring up the volume of the other frequencies making the s sounds less audible, others say by compressing the vocal you are bringing up the ss's along with everything else making them more audible. My own thoughts are I guess it all depends how loud [ not to the ear but in db terms ] the offending ss's are compared to the rest of the vocal. I would think if they were louder then compression may help by compressing them bringing the rest of the vocal up making the ss's less noticeable. However if the ss's are cutting through yet softer [ in db terms] compression may make them more noticeable by compressing the non-offending parts of the vocal and not the ss's. in case you have not noticed ss's drive me nuts. I've used several de-essers as well as multiband compression like Waves F6 to try and compress only the offending frequencies. Alas while I've been able to improve some situations there still seems to be no substitute for getting it right when you record it.
Compression and deessing
- Thread starter songsj
- Start date
Steenamaroo
...
I always find that those sorts of sounds are stronger after heavy compression.
I suppose if you look at the waveform of an offending S sound that part of the waveform is rarely as big as the body of the sound that follows.
Usually the compression is taming the strong body of sound and, therefore, the small noises like lip smacks, breaths, esses etc sound much louder.
I generally volume automate/clip gain automate anything like that which stands out.
I suppose if you look at the waveform of an offending S sound that part of the waveform is rarely as big as the body of the sound that follows.
Usually the compression is taming the strong body of sound and, therefore, the small noises like lip smacks, breaths, esses etc sound much louder.
I generally volume automate/clip gain automate anything like that which stands out.
andrushkiwt
Well-known member
Manually edit the gain first. "Esses" show as dense blobs in the waveform - easy to spot. More control over it's sound than a compressor too.
I haven't thought about this for a while -generally do gain edits on esses, but then I generally don't do much heavy vocal compression either.
How about just including an appropriate ess' eq bump into the sidechain of the vocal comp?
Realize a lot of our fav comps don't have sidechain access, (they ought to these days :>
.. but here's a cool tool for things like this.
Blue Cat's PatchWork - Fully Configurable Plug-Ins Chainer and Multi FX / Standalone Host or Plug-In
How about just including an appropriate ess' eq bump into the sidechain of the vocal comp?
Realize a lot of our fav comps don't have sidechain access, (they ought to these days :>
.. but here's a cool tool for things like this.
Blue Cat's PatchWork - Fully Configurable Plug-Ins Chainer and Multi FX / Standalone Host or Plug-In
Last edited:
ashcat_lt
Well-known member
Compression tends to favor low frequencies just because of the way it works and then - as mentioned above - the actual sibilance is never anywhere near as loud as the rest of the vocal. Unless you're EQing the sidechain pretty heavily or are really squashing the hell out of it, the comp only touches those s sounds accidentally - because it's still releasing from whatever came before. To be pedantic, it's not "bring the s sounds up", but bringing everything else down closer to where the sibilance was to begin with.
All that said I really think this is one of those things where we all tend to just get stuck. Like, you focus on those s sounds so you hear them everywhere and it seems like it's just ripping your head off, but that's often only because you've got it on the brain and have lost perspective. It's was probably fine a while ago and an objective comparison with similar material will show that you've probably gone too far by now.
Ideally, the whole thing is handled at tracking. Performance, placement, a little technique. Then you should almost be able to get away with simple static EQ.
A large part of the "problem" is that a typical mix has almost nothing else in that range. Cymbals, maybe the leading edge of other drum transients, and vocals. So they do tend to stand out simply because there's not much to mask them.
All that said I really think this is one of those things where we all tend to just get stuck. Like, you focus on those s sounds so you hear them everywhere and it seems like it's just ripping your head off, but that's often only because you've got it on the brain and have lost perspective. It's was probably fine a while ago and an objective comparison with similar material will show that you've probably gone too far by now.
Ideally, the whole thing is handled at tracking. Performance, placement, a little technique. Then you should almost be able to get away with simple static EQ.
A large part of the "problem" is that a typical mix has almost nothing else in that range. Cymbals, maybe the leading edge of other drum transients, and vocals. So they do tend to stand out simply because there's not much to mask them.
Steenamaroo
...
+1. This is a big deal for me.
I find if I focus on any given thing, I focus too hard.
For sibilance, lip smacks, guitar string squeaks etc I tend to go about my business doing other things, and just automate these things as they jump out at me.
It might sound silly but when I consciously make the decision to 'fix' all the sibilance in a track, I almost always go too far.
miroslav
Cosmic Cowboy
It's a lot of passes for me...back-n-forth.
When I'm comping a vocal track, I try to ignore all but the obvious *noises*...because I'm often in solo mode when comping.
Once I have the track...it's a lot of passes and flipping between solo and full playback before I find the right amount of adjustment.
Sometimes I'll pull the "S" stuff back on a phrase...then later I end up turning it back up. I've also had cases where I need to increase the "S" because it just didn't get captured loud enough during tracking...so the word looses meaning without it.
Also...I find that our brains will put in what's missing....IOW, it may seem bad during editing/mixing...but during final playback, it's just not that important for it to be perfect, because it works fine in the total sound.
That said...I do all my dessing manually, one by one, as they are all different, and sometimes a louder "S" works perfectly in the mix...so you don't want to mess with it. I've never used a comp/de-esser for that chore.
bouldersoundguy
Well-known member
I mixed a project that was recorded elsewhere and the singer was not available to retrack anything (illness), so processing was the only option. I thought I had it under control with de-essers, but when we took it to the (renowned) mastering engineer he described the vocal as "spitty" and I went back to the drawing board. I ended up manually editing the whole project (16 long songs), lead and harmony vocals. I didn't just edit S sounds, I edited Ts, Ds Ks, CHs, breaths and anything that didn't sound right. But I found that doing it in the mix helped me avoid over correcting. I mixed each sound to be right in the mix. It was a massive, mind numbing process in what was already a massive project, but it was worth it. And now it's our standard procedure for almost all vocal tracks.
Steenamaroo
...
Man, that says it all.
Huge pain in the ass and now I do it every time!
Huge pain in the ass and now I do it every time!
miroslav
Cosmic Cowboy
Yeah...I go through that every time, even when initially the vocals sound good, but as I listen over and over, I start to uncover the issues.
I'll do a rough scrub in solo mode....and then like a LOT of passes in full mix-mode to fine tune things and not over-fix shit that isn't a real mix issue.
I got so tired of trying to fix a couple of spots this weekend that I just went back since I'm the singer and re-sang two lines that had 4 s's in them I could not live with. I'm much happier now. I might have cooked the pre's a bit on the first one because no matter how soft I made them they just did not sound right. Too soft and it starts to sound like a lisp. This was my first go with a Tascam 16x8 USB interface so I was learning a lot at once. I will be much more careful in the future. I am also watching my s's when rehearsing. Should make me an easier singer to record. That said I'm 62 and kind of am what I am.