I got a revision request from a producer who thought the vocals were a little harsh in the 9 - 10 kHz area and he wanted them to sound a little smoother.
Now, harshness is generally found in the 2.5 kHz area, but you always want to consider the harmonic multiples as well, so it's not surprising that the 10 kHz area (2.5 kHz * 4) will add its own hissy-harshness (a new term I just made up).
I found three ways you can tackle this problem and I thought you might find them helpful.
Multi-Band Compression
You can use a multi-band compressor with only the top-band activated. That way the compressor leaves the entire signal alone, except for region above 9 kHz. Think of this as a "shelving compressor," another term I just made up. Man, I'm on fire. That's what a 5 am run to the new Post Malone album does to you!
De-Esser
A de-esser is really just a frequency-dependent compressor so it's not exclusively useful for sibilance. You can easily use it for a different frequency area by tweaking the area the de-esser should "hear." This would be the equivalent of having a compressor compress a bell-curved region instead of everything above a certain frequency.
EQ
Of course, the simplest solution is to just use an EQ. Find the offending frequency that's causing the excessive harshness and wipe it out.
