Sibilance is one of those problems that's obvious when it's wrong but invisible when it's right.
Too much sibilance? Your vocal sounds harsh, piercing, and fatiguing to listen to. Too much de-essing? Your vocal sounds like the singer has a lisp, with soft, mushy consonants that lack definition.
The goal isn't to eliminate sibilance—it's to control it so your vocal sounds natural, clear, and professional.
In this guide, we're breaking down how to de-ess vocals properly: where to place your de-esser in the chain, how to find your specific sibilance frequency, and how to avoid the most common mistakes that make vocals sound worse instead of better.
What You'll Learn
- What sibilance actually is (and why it's not always at the same frequency)
- Where to place your de-esser in your vocal chain (this might surprise you)
- How to find your specific sibilance frequency
- Common de-essing mistakes and how to fix them
- When to use two de-essers instead of one
What Is Sibilance (And Why Does It Vary)?
Sibilance refers to the harsh, high-frequency sounds produced by "S," "T," "SH," and similar consonants.
When you say "sibilance" or "sales" or "short," you're creating a burst of high-frequency energy—typically between 4kHz and 10kHz, depending on the singer's voice.
Here's what most people don't realize: sibilance frequency isn't universal. It varies based on:
- The singer's voice (male vs. female vs. child)
- The singer's natural tone and articulation
- The microphone used
- The proximity effect and mic technique
Common sibilance frequency ranges:
- Male vocals: 5kHz - 8kHz
- Female vocals: 6kHz - 10kHz
- Children's vocals: 7kHz - 12kHz
But these are just starting points. Your job is to find the specific frequency where sibilance is problematic for the vocal you're working on—not just assume it's at 7kHz because that's what a tutorial said.
The Problem: Most Producers Get De-Essing Wrong
If you've struggled with de-essing, you've probably experienced one of these issues:
Problem 1: You can't find the right frequency You set your de-esser to 7kHz like a tutorial told you, but the sibilance is still there. Or you're reducing the wrong frequency, making the vocal sound dull without fixing the harshness.
Problem 2: You over-de-ess and create a lisp You crank the de-esser to eliminate all harshness, and now your singer sounds like they can't pronounce their S's. The vocal loses clarity and definition.
Problem 3: You put the de-esser in the wrong place in your chain You de-ess early in your chain, then add an exciter or air band later—which re-enhances the very frequencies you just controlled. Now you're chasing your tail.
Problem 4: You confuse sibilance with brightness Sibilance is specific harsh bursts on consonants. Brightness is the overall tonal balance of the vocal. De-essing won't fix a vocal that's too bright—EQ will.
Here's the truth: de-essing is simple when you understand the principles. You just need to know where to place it, how to find your frequency, and when to stop.
Where to Place Your De-Esser in Your Vocal Chain
This is where most people get it wrong.
The common advice: Put your de-esser first in the chain to set up your other plugins for success.
The problem: If you add exciters, air bands, or high-frequency boosts later in your chain (which are super popular in modern vocal production), you're re-enhancing the sibilance you just controlled.
Now you've got two options:
Option 1: Move Your De-Esser Later in the Chain
Place your de-esser after any EQ, saturation, exciter, or air band processing.
This way, you're controlling sibilance at the final stage—after you've added brightness or air—so nothing downstream re-enhances those harsh frequencies.
Typical chain order:
Gain/trim
EQ (corrective)
Compression
EQ (tonal shaping)
Saturation/exciter (if using)
De-esser ← Place it here
Reverb/delay (on sends)
Option 2: Use Two De-Essers
If you need to control sibilance early and late, use two de-essers with light settings on both.
First de-esser: Early in the chain (after initial EQ, before compression) with gentle reduction (3-4dB max)
Second de-esser: Late in the chain (after saturation/exciter) with gentle reduction (2-3dB max)
This prevents over-processing while catching sibilance at multiple stages.
Why this works: Compression and saturation can emphasize sibilance. The first de-esser controls it before those processors make it worse. The second de-esser catches any harshness added by exciters or air bands.
How to Find Your Sibilance Frequency
Here's the step-by-step method:
Step 1: Solo the Vocal
Play a section with obvious sibilance (look for lines with lots of S, T, or SH sounds).
Step 2: Sweep with a Narrow EQ Band
Open an EQ, create a narrow band (high Q), and boost it by 10-15dB.
Sweep through the 4kHz - 10kHz range slowly while the vocal plays.
Step 3: Listen for Harsh, Piercing Sibilance
You'll hear different frequencies as you sweep:
Some will sound bright but pleasant
Some will sound harsh, piercing, and fatiguing—that's your sibilance frequency
Step 4: Set Your De-Esser to That Frequency
Once you've identified the frequency (let's say 7.2kHz), set your de-esser to target that exact frequency.
Step 5: Adjust Threshold and Reduction
Start with a low threshold so the de-esser only activates on the loudest sibilant bursts.
Gradually increase the threshold until you're catching most of the harshness—but not so much that the S's sound soft or lisped.
What you're listening for:
Sibilance should be controlled, not eliminated
Consonants should still sound crisp and clear
The vocal should feel smooth but not dull
Common De-Essing Mistakes (And How to Fix Them)
Mistake #1: Using the Default Frequency Without Checking
Most de-essers default to around 7kHz. That's fine as a starting point, but it's rarely the exact frequency you need.
The fix: Always sweep to find your sibilance frequency. Every voice is different.
Mistake #2: Over-De-Essing and Creating a Lisp
You crank the de-esser to eliminate all harshness, and now the singer sounds like they can't pronounce their S's.
The fix: De-essing should be subtle. You're controlling sibilance, not removing it. If you're reducing by more than 6-8dB, you've gone too far.
How to check: A/B your de-essed vocal with the original. The S's should sound smoother but still present and clear.
Mistake #3: Placing the De-Esser Too Early in the Chain
You de-ess first, then add an exciter or air band, which re-enhances the sibilance.
The fix: Move your de-esser to after any high-frequency enhancement (saturation, exciter, air bands). Or use two de-essers—one early, one late.
Mistake #4: Confusing Sibilance with Brightness
Your vocal sounds harsh overall, so you crank a de-esser to tame it. But the harshness is still there, just duller.
The fix: De-essing targets specific bursts of high-frequency energy on consonants. If your vocal is too bright overall, use EQ to reduce the high end—don't rely on a de-esser.
Mistake #5: Not Checking in Context with the Mix
Your vocal sounds perfect in solo, but in the full mix, sibilance is cutting through harshly.
The fix: Always check your de-essing in context with the instrumental. Sibilance that sounds fine in solo can be piercing in a dense mix.
When to Use Two De-Essers
If you're adding saturation, exciters, or air bands to your vocal, a single de-esser might not be enough.
Use two de-essers when:
You're adding high-frequency enhancement later in the chain
The vocal has extreme sibilance that needs control at multiple stages
Compression is emphasizing sibilance (compress, then de-ess after compression)
How to set them:
First de-esser: Target the primary sibilance frequency (e.g., 7kHz), reduce by 3-4dB max
Second de-esser: Target the same frequency or slightly higher (e.g., 8kHz), reduce by 2-3dB max
This prevents over-processing while maintaining control throughout your chain.
The Hidden Problem: Hi-Hats and Sibilance Frequency Overlap
Here's something most producers don't realize: hi-hats and sibilance often occupy the same frequency range (6kHz - 10kHz).
When a hi-hat hits at the exact moment the singer says an S, the two sounds blend together. The result? It sounds like the singer didn't say the S clearly, or the hi-hat becomes piercing and harsh.
The solution:
EQ the hi-hat, not the vocal - Use a narrow EQ cut on the hi-hat to separate it from the vocal's sibilance frequency
Pan or time-shift - If the hi-hat and sibilance are consistently clashing, slightly adjust the timing or panning of the hi-hat to create separation
Sidechain compression - Use the vocal to duck the hi-hat slightly during sibilant moments (advanced technique)
This is especially common in electronic music, where producers mix their own tracks and brighten vocals aggressively—creating piercing sibilance that fights with percussion.
Polarizing Take: Electronic Producers Overdo Vocal Brightness
A lot of electronic producers mix their own vocals, and they tend to brighten them way too much.
Why? Because electronic production often has a lot of high-frequency content (hi-hats, synths, digital percussion), and producers compensate by making the vocal even brighter to cut through.
The result: vocals that pierce your ears with harsh, fatiguing sibilance.
If you're an electronic producer, here's the hard truth: your vocal doesn't need to be brighter than everything else to sit in the mix. It needs to be clear and present, which is different.
Try this instead:
Use less high-frequency boost on the vocal
De-ess properly (not excessively)
Carve space in the instrumental (EQ dips in synths/pads at vocal frequencies)
Use saturation for presence instead of raw brightness
Your ears (and your listeners' ears) will thank you.
Do You Always Need to De-Ess?
Not always.
Skip de-essing when:
The vocal has naturally smooth sibilance (some singers just don't have harsh S's)
The genre calls for a raw, unpolished sound (punk, lo-fi, some indie)
You've already controlled sibilance with mic technique (proper distance, off-axis positioning)
Always de-ess when:
The vocal is harsh or fatiguing on headphones
Sibilance is cutting through the mix too aggressively
You're adding exciters or air bands (which will emphasize sibilance)
Sibilance vs. Brightness: What's the Difference?
This is where a lot of people get confused.
Sibilance: Short bursts of high-frequency energy on specific consonants (S, T, SH, CH, Z). It's transient—it comes and goes with certain syllables.
Brightness: The overall tonal balance of the vocal across the entire frequency spectrum. If a vocal is too bright, it means the high end (5kHz - 15kHz) is too loud constantly, not just on consonants.
How to fix each:
Sibilance → Use a de-esser (dynamic, transient-based reduction)
Brightness → Use EQ (static, overall tonal reduction)
If you use a de-esser to fix brightness, you'll end up over-processing the consonants while the rest of the vocal stays too bright. That's why it sounds weird.
Tools: Dynamic EQ vs. Multiband Compression vs. De-Esser
You can control sibilance with different tools. Here's when to use each:
Traditional De-Esser
Best for: Simple, straightforward sibilance control
How it works: Detects high-frequency peaks and reduces gain dynamically
Pros: Easy to use, fast workflow
Cons: Less surgical control
Dynamic EQ
Best for: Precise, surgical sibilance control
How it works: Only reduces the exact frequency when it exceeds a threshold
Pros: More transparent, less side effects
Cons: Requires more setup and fine-tuning
Multiband Compression
Best for: Controlling sibilance as part of overall high-frequency dynamics
How it works: Compresses the high-frequency band (e.g., 5kHz+) dynamically
Pros: Can smooth out harsh high end overall
Cons: Less precise, can affect non-sibilant high frequencies
My recommendation: Start with a simple de-esser. If you need more control, graduate to dynamic EQ. Save multiband compression for broader tonal shaping, not just sibilance.
Common Questions About De-Essing
How much reduction is too much?
If you're reducing by more than 6-8dB on the loudest sibilant peaks, you're probably over-processing. The goal is to smooth out harshness, not eliminate consonants entirely.
Should I de-ess before or after compression?
Both can work, but after compression is often better. Compression can emphasize sibilance, so de-essing after compression ensures you're controlling the final result.
Can I use a de-esser on instruments?
Yes! De-essers work great on acoustic guitars, hi-hats, cymbals, and any instrument with harsh high-frequency transients. Use the same principles: find the harsh frequency, set a threshold, reduce gently.
Do I need an expensive de-esser plugin?
No. Most DAWs include a basic de-esser that works fine. What matters is how you use it—finding the right frequency, setting the right threshold, placing it correctly in your chain.
What if my vocal still sounds harsh after de-essing?
You might be dealing with brightness, not sibilance. Try a gentle high-shelf EQ cut (starting around 8kHz) to reduce overall harshness. Or check if there's resonance or honkiness in the midrange (2kHz - 4kHz) that's making the vocal sound harsh.
Final Thoughts
De-essing isn't glamorous, but it's essential.
The difference between a harsh, fatiguing vocal and a smooth, professional one often comes down to proper sibilance control.
You don't need expensive plugins or complicated techniques. You need to:
Find your specific sibilance frequency (sweep with EQ)
Place your de-esser correctly in the chain (after exciters/air bands)
Reduce gently (control, don't eliminate)
Check in context with the full mix
Start with one vocal. Sweep to find the harsh frequency. Set your de-esser. Reduce until the harshness is smooth but the consonants are still clear.
That's how professional vocal mixes are built—one intentional adjustment at a time.