First thing, see if she can record a bit further back from the mic.

 

What I always tend to do is to watch the spectrum of the voice on an analyzer and watch where the problem lives and then threat it with a deesser.

 

A small example of how a good vocalist can sing into a large diaphragm condenser WITHOUT popfilter, and no sybillants.
The mic is a Brauner VMX, a rather bright sounding mic. Not as bright as VM1, but still...

www.youtube.com/watch?v=04l_bmp3y4Q

 

Last edited: Jan 18, 2017

Mostly everything that needs to be said has been, but I'll just add that when it comes to trying plugin solutions give SKnote SoundBrigade a shot. I use this in tandem with fabfilter DS when I have to tame serious sibilance issues.

 

Last edited: Jan 18, 2017

There 2 types/modes of DE ESSERS.

multiband: this is basiclaly a multiband compressor. If the S is dectected in the region (2-7k or whatever), then only that band is turned down.

broadband: If the S is dectected in the region (2-7k or whatever), the WHOLE sound is turned down, might sound more natural.

Read docs on which one(s) your implementation(s) supports and play with settings.

 

I do get asked a lot about de-essers in general.
sibilance is actually something that is a part of the voice,


The thing is de-essers are compressors in essence so you got to pay attention to the amount of reduction
and the threshold.
set the threshold to low and it doesn't only reduce the sibilance, but much more and that will defeat it's purpose.

I also wouldn't re comment de-essing on a single pair of headphones, most headphones are louder at the sibilant frequency's.