De-essing is the process of reducing harsh, piercing “S”, “SH”, “T” and “Z” sounds (known as sibilance) in a vocal recording, so the voice sits smoothly in the mix instead of stabbing the listener’s ears. The quickest way to de-ess vocals is to add a de-esser after your EQ and compression, set it to listen to the sibilant band (usually somewhere between 5 and 9 kHz), then lower the threshold until the harshness disappears without turning the singer into a lisp. In this 2026 guide you will learn what sibilance really is, how to find the exact frequency on any voice, the de-esser settings that work on most singers, the mistakes to avoid, and when it pays to hand the mix to a professional studio.
Table of Contents
- What is de-essing and sibilance?
- Why vocals get harsh “S” sounds
- How to de-ess vocals: step by step
- De-esser settings that work on most voices
- De-essing methods compared
- Common de-essing mistakes
- When to hand it to a pro
- FAQ
What Is De-Essing and Sibilance?
Sibilance is the burst of high-frequency energy your voice produces on consonants like S, SH, CH, T and Z. It is a natural part of human speech, but a microphone – especially a bright large-diaphragm condenser like the Neumann U87 we use at G Studio – can exaggerate it. What sounds normal in the room can turn into a sharp, whistling “ess” once it is recorded, compressed and brightened in a mix.
Sibilant consonants concentrate most of their energy between roughly 4 kHz and 10 kHz. De-essing is simply a fast, targeted volume reduction that clamps down on that band only when a harsh “S” appears, and otherwise leaves the voice untouched. Think of it as a very quick, very narrow compressor that is triggered by sibilance alone.
Why Vocals Get Harsh “S” Sounds
Harsh sibilance rarely comes from one place. Usually it is a stack of small causes:
- Microphone choice and distance. Bright condensers and singing too close to the capsule push up the high end.
- Mic technique. A singer who points every “S” straight at the diaphragm will record more sibilance than one angled slightly off-axis.
- Compression. Heavy compression raises the quiet detail in a voice, and sibilance rides up with it.
- High-shelf EQ. Adding “air” at 10-12 kHz for sparkle also lifts the “S” region.
- Saturation and exciters. Anything that adds harmonics adds high-frequency content too.
This is why de-essing almost always comes after compression and tone-shaping: you are cleaning up the sibilance those stages introduced. Good vocal recording technique reduces the problem at the source, but even a perfect take usually needs a touch of de-essing in the mix.
How to De-Ess Vocals: Step by Step
1. Place the de-esser correctly in the chain
Insert the de-esser after your corrective EQ and main compressor, but before any bright high-shelf EQ, reverb or delay. If you add air first and de-ess afterwards you will fight yourself, and if reverb comes first the de-esser cannot catch the sibilance in the tail.
2. Find the sibilant frequency
Set the de-esser to its “listen” or “solo” mode and sweep the detection frequency up and down while the singer sings a line full of S sounds. You are hunting for the spot where the “ess” is loudest and most annoying. On most voices it lands between 5 and 9 kHz: deeper male voices sit lower (around 5-7 kHz), brighter female and child voices sit higher (7-9 kHz).
3. Set the threshold
Turn the “listen” mode off and lower the threshold until the meter shows gain reduction only on the sibilant syllables – not on the vowels or the body of the voice. You want the de-esser resting at zero most of the time and dipping 3-6 dB on the worst “S” sounds.
4. Choose wideband or split mode
Split-band (or “frequency-specific”) mode only ducks the high frequencies, which keeps the voice natural. Wideband mode ducks the whole signal when sibilance hits, which can sound more transparent on gentle sibilance but may cause a slight volume dip. Start with split-band for most pop and R and B vocals.
5. Check by bypassing
Toggle the de-esser on and off while listening to the full mix, not the solo vocal. The goal is a voice where you stop noticing the “S” sounds – not one that suddenly lisps. If the singer sounds like they have a cold, you have gone too far.
6. Automate the stubborn words
Some words will spike no matter how you set the plug-in. Rather than cranking the de-esser harder across the whole song, draw a short volume automation dip on just those words, or use clip gain to pull down the single harsh syllable. This keeps the rest of the vocal open and alive.
De-Esser Settings That Work on Most Voices
These are safe starting points. Always trust your ears over any numbers – every voice, mic and room is different.
| Voice type | Detection frequency | Gain reduction | Mode |
|---|---|---|---|
| Deep male vocal | 5-7 kHz | 3-5 dB | Split-band |
| Bright female vocal | 7-9 kHz | 3-6 dB | Split-band |
| Child or very bright voice | 8-9 kHz | 2-4 dB | Split-band |
| Rap / spoken word | 6-8 kHz | 4-6 dB | Wideband or split |
De-Essing Methods Compared
A dedicated de-esser is the fastest tool, but it is not the only way to tame sibilance. Here is how the main approaches compare:
| Method | Best for | Effort | Transparency |
|---|---|---|---|
| Dedicated de-esser plug-in | Everyday vocals, fast results | Low | High |
| Dynamic EQ | Precise, surgical control of one band | Medium | Very high |
| Multiband compressor | Sibilance plus overall tone control | Medium | Medium-high |
| Manual clip gain / automation | A few stubborn words | High | Highest |
In a professional mixing and mastering session we usually combine two: a light de-esser across the whole vocal to handle the average, plus manual automation on the handful of words that still poke through. That combination keeps the voice smooth without dulling it.
Common De-Essing Mistakes
- Over-de-essing. The most common error. Too much reduction removes the natural crispness and the singer sounds like they are lisping.
- De-essing in solo. Sibilance that is obvious on the soloed vocal often disappears in the full mix. Always make the final decision in context.
- Wrong frequency. If the de-esser is set too low it will clamp down on vowels and make the voice sound muffled and dark.
- One de-esser for everything. A busy, dense vocal performance may need two gentle de-essers in series rather than one aggressive one.
- Fixing it only in the mix. If every take is harsh, the real fix is mic choice, position and a pop/sibilance-friendly technique during recording.
When to Hand It to a Pro
De-essing is one of those tasks that looks simple and takes years to get consistently right. The difference between a vocal that merely has its “S” sounds removed and one that sounds effortless and expensive is experience – knowing exactly how much to pull back before the life leaves the voice.
At G Studio on Palm Jumeirah, de-essing is a small part of a full vocal mix. We record through a Neumann U87 and a Tube-Tech MP2A chain, which already keeps sibilance under control at the source, and then balance de-essing against compression, EQ and reverb so the voice sounds polished rather than processed. If you have tracked a song and the vocals feel harsh, a professional mix is often the fastest path to a release-ready sound. You can also compare this with our guide on how to compress vocals, since compression and de-essing work hand in hand.
Frequently Asked Questions
What is sibilance in vocals?
Sibilance is the sharp burst of high-frequency energy produced on consonants like S, SH, CH, T and Z. It is normal in speech but can become harsh once a voice is recorded, compressed and brightened, which is why mixers de-ess it.
What frequency should I de-ess at?
Most sibilance lives between 5 and 9 kHz. Deeper male voices usually sit around 5-7 kHz, while brighter female and child voices sit around 7-9 kHz. The best method is to sweep the detection band and listen for where the “S” is loudest.
Should I de-ess before or after compression?
Usually after. Compression tends to raise sibilance, so de-essing afterwards cleans up what the compressor exaggerated. Place the de-esser before any bright high-shelf EQ, reverb or delay.
Can you de-ess too much?
Yes. Over-de-essing removes the natural crispness of a voice and can make the singer sound like they are lisping or have a cold. Aim for 3-6 dB of reduction only on the harshest syllables, and always check in the full mix.
Do I still need a de-esser if I used a good microphone?
Often, yes. A great mic and good technique reduce sibilance, but compression and EQ later in the chain can still bring it back. Even professionally recorded vocals usually get a light touch of de-essing.
How much does professional vocal mixing cost in Dubai?
At G Studio, studio time is AED 450 per hour with an engineer and AED 600 per hour with George himself. De-essing is included as part of a full vocal mix rather than charged separately. For an exact quote, send us your track and we will advise.
About the Author
George Yukhanov is a multi-platinum, award-winning songwriter, composer and producer with more than 20 years in music and over 500 million views and 50 million streams across his work. He ranks among the top 5% of producers worldwide and is one of the rare multi-platinum producers who works openly with newcomers. He founded G Studio, a premium recording studio on Palm Jumeirah in Dubai, where artists at every level record, produce and mix release-ready songs.