If you already know what de-essing is, skip to the second paragraph.
Sibilant- a type of fricative or affricate consonant, made by directing a jet of air through a narrow channel in the vocal tract towards the sharp edge of the teeth.
Basically, the "s" sound. Ever hear a recording where the "s"s and "t"s seem really loud compared to rest of the vocals. That because the vocalist is pushing out the same amount or more air through a smaller opening toward the mic. Put your mouth in the "ah" position and then make it like you're ready to say "tuh". It's easy to tell that the opening of your mouth gets smaller and therefor the gush of air that will come out will be stronger, just like if you put your thumb over half the opening of a hose. Same thing may happen when the vocalist say a "p" word. To make the "p" sound, you gather air in your mouth and release it all at once. So if the diaphragm of the mic is receiving these strong gusts of air, it will record it as being louder than the rest of the vocals. For the "s" and the "t", the pitch that's created is higher than normal vocals and the "p" will be lower. The "p" is the low pitch burst you hear is some recordings.
So what can you do to get rid of these outlandish sibilants and pop? Pop filters are a must. You can make one with a pair of nylons and a coat hanger that rival professional pop filters. Pop filters are great but they probably won't get rid of everything. There are also plug-ins (I'm sure there's a VST) that you can simply put on the vocals channel in the mixer and it will take a guess at what needs to be fixed and fix it. I'm going to go through the steps of how to make a custom de-esser, without purchasing and extra plug-in. This will also give you greater control over what you are fixing as apposed to presets. Also, most people on WB are probably using FL Studio, so I'm going to go through how to make it on that DAW and tell you how you can probably figure out how to do it on another DAW which has actual busses for signal routing as apposed to FL's simple channel routing.
1) To set it up, copy your vocals and link each copy to a different channel in the mixer. I'm going to use 1 and 2. Channel one will contain the actual vocals and channel 2 will contain the one used for de-essing. In channel one, insert a compressor and set it up how you like it for the vocals. Not too much but not so little that it has no effect. The attack should be 0.0ms with a release of what ever sounds good in the mix.
2a) Insert a parametric EQ on channel 2.
2b) For FL users: Click the Mute/Solo button above the channel's pan nob so the green light shuts off. OK, now turn it back on. haha. You just need to know where it's at for after step 3. Route channel 2 to channel 1 by selecting channel two and clicking the up arrow on channel 1 that's just above the orange box that says "FX" near the bottom of the insert. This will make channel 2's signal go through channel one before going through the Master channel. Also, the nob that appears when you do this controls the amount of signal that goes from channel 2 to channel 1. You'll need to know this later.
2c) For non-FL users: You don't need to worry about muting channel 2 if you are using busses because the signal of channel 2 will never reach the Master. AFTER step 3, select an unused mono bus as the output of channel 2. Most compressor plug-ins will have a Key Input. In the compressor on channel 1, click on the "Key Input" button and select the bus that you selected as the output of channel 2. This is also after step 3. Sorry if this is out of order, it just fit in with whats happening in FL at this point.
3) The fun part. Open up the Parametric EQ on channel 2. Well maybe I should say this first. Everyone's voice is different, that's why I'm telling you this technique as opposed to telling you to invest in a D-esser plug-in. The frequency that your sibilants come out are different than, say Shaun Connery's sibilants. His "s"es are remarkably low pitch in comparison to most peoples'. Say "ess" in your voice then do a Shaun Connery impression and say "ess". Much lower pitch right? That's why you want the control this technique gives you over the standard D-esser plug-in that may only have one setting! The point is, most peoples' sibilants come in the frequency range at around 5-8kHz. This is a good starting range. Play the vocals that are connected to channel 2 as you're doing this. Have a part with a lot of "s"s and "t"s playing on repeat. In the EQ, make a high-pass filter that cuts out everything from AROUND 5kHz and down. This is what I'm talking about with control. 5kHz is a starting point. Find the frequency that really make those "s" and "t" sounds extremely exaggerated. The louder and more obnoxious the better. Cut out all frequencies of the actual vocal except for the sibilants. And boost the sh!t out of those "s"s and "t"s. You may want to save this Parametric EQ setting and make multiples of it on channel 2 to make sure you're cutting out all of the vocals and boosting all the of the sibilants. For other DAWs, just command click and drag down another copy of the EQ onto the channel. If having the compressor boosting those frequency maxes it out and creates clipping and/or distortion, tone down the boost and just make more copies of the EQ on channel 2"
*Remember parts 2b and 2c. In FL, click the green Mute Solo so it's on and for other DAWs, do your channel bussing and set up channel 1's compressor Key Input.
4) Play both vocals. Should be the same vocals just one each of the two channels. If everything was set-up right, the extra sibilants from channel 2 are sent through the compressor on channel 1 and creating extra compression ONLY when the singer says his "s"s and "t"s too loud. Also, the volume signal of channel 2 should also be coming through but NOT playing actual sound through the Master channel. This is where that nob above the channel router comes in in FL. Turning it up will create more compression and turning it down will create less. Just have to listen. Turning it up too high will make your vocalist sound like he has a lisp. Turn it down too far and it's not doing anything. In other DAWs, just adjust that one compressor on channel one. That compressor should be devoted just for sibilants and you'll want to add an extra compressor if you want normal level management on that vocal channel.
Remember to open your mind when you're reading or watching tutorials like this. Just because this is most commonly used on vocals doesn't mean it's the only application. Same goes for any technique you learn. Just because it's used one way doesn't mean you can't mess with it and discover something that sounds great. Just think of whats actually happening to your sound instead of just what the outcome is that one time.
Hope this helps.