If you’ve ever pulled up a vocal take and winced every time the singer hit an “s” or a “sh,” you already know what sibilance is, even if you didn’t have a name for it. In this guide, I’ll walk you through what de-essing actually is, how to use a de-esser, and how the process looks in practice using De-Sipper, which is $20 de-esser plugin from Carbonated Audio.
What Is De-Esser in Music?
A de-esser is a compressor that only targets one narrow frequency range, usually between 4kHz and 10kHz, where sibilant sounds like “s,” “z,” “sh,” and “ch” tend to sit. Instead of squashing the whole vocal the way a normal compressor does, it listens for that specific range and clamps down on it only when it spikes, then lets everything else pass through untouched.
This is different from just rolling off high end with EQ. A static EQ cut removes brightness all the time, whether sibilance is happening or not. A de-esser only reacts when the sibilance actually spikes, so the vocal keeps its natural air and presence the rest of the time. That reactive behavior is the whole point of the tool.
Sibilance itself usually comes down to a couple of things working together, and it’s worth knowing both before you ever touch a plugin.
- Mic technique and proximity
Condenser mics are naturally sensitive in that same 5kHz to 9kHz zone, so they pick up every bit of sibilance a vocalist produces, sometimes more than what you’d hear standing in the room with them.
The closer the singer is to the capsule, the more exaggerated this tends to get, which is why two takes from the same person can sound completely different depending on mic distance alone..
- EQ boosts during mixing
If you boost presence or air to make a vocal sound brighter or more “radio ready,” you’re often making the sibilance worse at the same time, since you’re amplifying the same range the “s” sounds live in. That’s exactly why de-essing typically happens after EQ, so the de-esser reacts to your final tonal balance instead of guessing at a problem that hasn’t fully formed yet.
While vocals are the most common use, I’ve also reached for a de-esser on harsh cymbals and on dialogue tracks where a host’s lisp got exaggerated under a sensitive mic. Anywhere a frequency range gets randomly too loud and too sharp, this tool can smooth it out without you riding a fader by hand.
How to Use a De-Esser on Vocals Using De-Sipper by Carbonated Audio
No matter which plugin you’re using, the basic order of operations stays the same: insert the de-esser after EQ and before your main compressor, find the frequency where the sibilance is happening, and set a threshold that only reacts to the harsh “s” and “sh” sounds, not the full vocal.
Here’s exactly how that plays out using De-Sipper, which runs as VST3, AU, or AAX on macOS and VST3 on Windows, so it works fine in Pro Tools, Logic, Ableton, Studio One, Cubase, and Reaper. It’s a $20 one-time purchase, no subscription, no iLok.
1. Insert it after EQ
The most important thing about De-Sipper is that it uses split-band processing, meaning it isolates just the sibilant frequency range and leaves the rest of the signal completely untouched.
This is the main reason it stays transparent instead of sounding squashed when you push it harder than you would with a more generic plugin. I’d drop it in right after your EQ stage so it’s reacting to the tone you’ve already shaped, not guessing at a vocal that’s about to change.
2. Use the S-Chain mode to hear what’s being caught
De-Sipper has a Monitor switch with two settings: Audio and S-Chain. Audio is your normal processed output, but flipping to S-Chain lets you solo just the sibilance the plugin is detecting, before you commit to any reduction amount.
Use this first, so you actually hear what’s being flagged instead of sweeping a threshold blind and hoping for the best. It only takes a few seconds and it’ll save you from chasing the wrong frequency for the rest of the session.
3. Move the threshold in small steps
When it comes to threshold, it ranges from subtle smoothing to a much heavier clamp, and I always start light and add more only if needed.
Over-de-essing is just as noticeable as under-de-essing, except instead of a harsh “s” you end up with a muffled, lispy one that’s honestly harder to fix afterward..
I also recommend you to bypass the plugin every few seconds to A/B against the dry signal as you go, since your ears adjust fast and you can lose perspective if you don’t check.
Watch the gain reduction
De-Sipper shows gain reduction and where it’s happening in real time. If the meter moves on words that don’t have any sibilance in them, that’s your cue to narrow the frequency target rather than just lowering the threshold and hoping it behaves.
De-Sipper also comes with a free demo that runs the full DSP with a 60-second-on, 10-second-muted cycle, so you can test it on real material before buying anything. It also comes with a 30-day money-back guarantee, so there’s not much risk in trying it on a session you’re already working on. You can grab it directly from the Carbonated Audio De-Sipper page.
Common Mistakes to Avoid
- Setting the threshold too aggressively
This is probably the most common one I see. It feels like progress when the meter shows a lot of gain reduction, but if you push past what the vocal actually needs, you end up with something dull or slightly lisped instead of clean. The fix is just patience: start light, listen back, and only add more if the harshness is genuinely still there.
- Placing the de-esser before EQ
If you de-ess before you’ve finished shaping the tone with EQ, you’re reacting to a balance that’s about to change anyway. Any presence or air boost you add afterward can reintroduce sibilance the de-esser already thought it handled, so you end up doing the work twice for no good reason.
- Skipping the dry and wet comparison
It’s easy to get used to the processed sound after a few minutes and lose track of how much you’re actually removing. Bypassing the plugin every so often keeps you honest about whether you’re still hearing the vocal’s natural character or whether you’ve quietly carved too much of it away.
- Using one setting across an entire song
Sibilance usually shifts between a quiet verse and a louder chorus, especially if the vocalist’s energy or distance from the mic changes between sections. A setting that works perfectly on the verse can undershoot or overshoot once the chorus hits, so it’s worth checking your settings against the loudest and quietest parts of the take, not just the part you happened to set it on.
Wrapping It Up
De-essing is just frequency-specific compression aimed at harsh “s” and “sh” sounds. The goal is for it to work quietly in the background, not announce itself every time it kicks in.
Find the frequency, start with a light threshold, listen closely, and only push harder if the problem is still there. De-Sipper’s free demo makes it easy to test that process on your own vocal before spending anything.
Music Circuit covers tech news, plugin reviews, and tutorials for musicians and producers. All gear is independently tested and reviewed with no sponsored influence on our ratings.
