The image to the right is a man saying “That’s scary.” In the middle is a big black bump where the t’s and sc all run together to make a big loud white noise in around the 3-6 KHz range.
If you EQ that range down you will be filtering out all sounds in that frequency, like hi-hats, finger snaps, the shimmer on the Rhodes, not to mention ruining that trippy reverb on the sample. What you really want is to dampen the range down but only when it gets above a certain volume — which is the definition of compression.
The problem with ‘S’ and ‘SH’ is so prevalent that audio engineers have a special class of tools that take the ‘s’ away — de ‘s’ — the original wave. These DeEssers are really just specialized compressors. The purple part of the wave is what was left after I ran the wave through Wave’s DeEsser with the threshold cranked real low — about -45.
If you are doing semi-professional work with samples, vocalists or “found” sounds you should probably pop for a pro-plugin package (like Waves) that has a great dynamics processing (EQ, compression, limiters, gates, etc.). Learning to use them is not that hard (even if learning to master them can take a really, really long time).
But the price for these things are not for the faint.
Many mixers and wave editors come with some kind of “multi-band compressor” and those should have a preset (example) that has some kind of hint about de-essing. These should do in a pinch.