is there a "best" way to De-ess vocals ???

Punkyskunky

Engineer at Brett Brothers Studio
Mar 8, 2014
82
6
8
Turkey
www.thomasbrettmixing.com
Hey guys, I'm interested to know, what technique do most of you guys prefer when De-essing vocals?

Until recently I was manually riding the volume for each sibilant sound in my vocals, but lately I've been using Fabfilter Pro-Ds simply because it's a lot less of a hassle... I've been using it in the wide band setting with 4x oversampling to reduce the distortion.

I guess my real question is: which method do you guys think is the most natural way to de-ess a vocal? Automating each sibilant letter? using a wide band de-esser? or simply using a split band de-esser (which i've noticed is what alot of people on the forum use)?

Seems like automating each sibilance manually would be the most natural sounding method to me, but I'm interested in what you guys think.

I'm not really looking for a definite "this is the best way!" answer, I'm more interested in why you guys use the methods you use...

Thanks for any replies in advance :D
 
For some reason in every situation I've recorded myself or anyone else, this problem has never come up even with multiple types of mics.
 
I wasn't trying to come off like a perfect asshole :lol: just that I've never had really sibilant vocals happen in recordings I've made. It's probably too late for something like that but I'd say try a different mic and maybe tell the singer to back off the mic a bit? I dunno, I'm just not really clear on how errant sibilance happens (room/mic/person etc.).
 
I wasn't trying to come off like a perfect asshole :lol: just that I've never had really sibilant vocals happen in recordings I've made. It's probably too late for something like that but I'd say try a different mic and maybe tell the singer to back off the mic a bit? I dunno, I'm just not really clear on how errant sibilance happens (room/mic/person etc.).

I wasn't being sarcastic :) , if you have near to no sibilance you honestly must be doing something right! For some reason all of the singers I record require quite a bit of de-essing, it might be because of the mic I'm using as you said... Unfortunately I can't get a new mic anytime soon :(
 
Most of the time a simple De-Essing plugin works for what I need, but in one particular instance I remember having to lower the volume of the "S" sounds a couple dB with automation in order to get the vocals to sit right.
 
I like waves renaissance Desser, both in wide and narrow mode. Sometimes i use 2 dessers if there are several problem areas. Also multibandscompressors works very well, just set the release time so the vocals breath and don't dull up to much.
 
I use stock PT de-esser because I don't have any other stuff + automation. I almost everytime narrow cut a lot around 8kHz and slightly boost around 13kHz. I've always read that basic vocal EQing includes a slight boost around 8Hz for it sounds harsh to me and simply does not work to my ears.
 
Been searching this forum for De essing tips. Mixing this band's stuff where the screams are perfectly fine, but the clean singing is insanely Essy. One of the lyric lines is something like "...that I surpass these thoughts of..." sang on one held out note. One big Ess fest. Even the "Th" and before smashing it with a compressor it was way essy. The more and more I tried to fix it the more and more it stuck out to me. After a while that's all I could hear hahaha. Anyways, after reading a few things I seemed to have fixed it by putting a de esser at the most prominent "S" sound around 6700hz then automating the volume of the esses down briefly a few db.
 
One of the very first editing tricks I've came up with and I use it ever since... I open the vocal take in an editor and go through the take and boost/cut most phases to roughly the same level (it was before I even know compressors exist) briefly listening to check if it sounds natural. De-essing is a side product of this approach too. It is actually faster than to setup compressors/de-essers and stuff to sound right.
 
Manually lowering the level on them is what I do, then just a general deessing is all that is needed. Clip gain in protools is great for this.
 
Usually I hate to use a desser, probably because I dont know how to use it properly, because every time I use it, it sucks all the clarity and brightness in the vocal track, even tweaked to only work with really high frequencies. So, doing volume automation or even eq automation it´s always better to my ears.
 
automation is the "best" way as its the only way you'll get it EXACTLY how you want it. I'm usually using 2 de-essers in different parts of the chain which works most of the time, otherwise its time to draw automation in....... after a while you recognise what "sssss" "fffff" "p" sounds all look like and it gets faster.
 
Thanks for all the great replies fellas!

I've been trying some different de-essing methods this week, and I think so far my fav. has been manually reducing each sibilant with automation, then using an instance of Fabfilter Pro-Ds in split mode at the very start of my vocal chain, and then another instance of Fabfilter Pro-Ds at the end of my chain if necessary.