Going along with my vocal editing basics article I wrote the other day, I thought I’d write a quick something about de-essing since I’ve used it in just about every vocal edit I’ve ever done. Especially when it comes to non-music applications.

De-Essing is frequency specific compression. It’s used for taking and extremely sibilant sound and getting rid of the frequencies that make the “s” sound so harsh. Pretty much every DAW I’ve encountered has some sort of stock de-esser (like in Pro Tools), but there are a lot of other ones, too. Take the Sonnox “SuperEsser,” for example.

Above is the De-Esser/Dyn 3, the stock de-esser inside Pro Tools 8. It doesn’t look too intimidating, right?

The first thing you need to do is decide if the vocals you’re editing sound harsh enough to warrant de-essing. I can’t really help you with that as it is pretty subjective. But a general rule of thumb: if it sounds bad, hurts your ears, or is distorting your master, you need to do something about it.

The de-esser has an in/out meter, a gain reduction meter, a listen option, a frequency selecter, and a “range” knob which will act as your threshold of sorts.

Play your audio back and hit listen. This will focus in on the frequency you have selected below. Think of it like what you do when you’re doing subtractive EQ: narrow band, turn the gain up so you can really hear all the boxiness/garbage that’s inside your audio. Find the frequency(ies) that’s most offensive. Now it’s time to turn them down!

In my particular application, 6.6kHz was “the zone.” The guy that recorded these V/Os is more sibilant than most, so it took me a little extra time to figure out his particular voice.

I turned the range knob up to -5.6dB, resulting in around 4-5dB of gain reduction. Remember: the point of de-essing is NOT to eliminate all sibilance, but to rather tame otherwise wild sibilance in your recording.

I was happy with these settings and it sounded much more listenable. Make sure you A/B your audio before comitting, though.