De-Esser Online: Remove Sibilance Without Dulling the Vocal

This comes up constantly, and the answers online are usually either a marketing page or a forum argument from 2014.

This covers De-Esser online the way it actually works in a session: the mechanism, the numbers that matter, and the decisions that follow from them.

Nothing below is theory for its own sake. Every number here is one you can check on a meter, and every step is one you can run on the next track you open.

Key takeaways

Chain order: high-pass, subtractive EQ, de-ess, compress, compress again, additive EQ, saturate.
Two gentle compressors beat one working hard.
Carve the instrumental to make room rather than pushing the vocal louder.
Set the key correctly before any pitch correction.

The vocal chain, in the order that works

Vocal processing is not a stack of favourite plugins. It is a sequence where each stage cleans up the input for the next one.

Stage	What it fixes	Typical setting
High-pass	Rumble, plosives, proximity buildup	80 to 120 Hz depending on the voice
Subtractive EQ	Boxiness and resonances	Narrow cuts at 200 to 500 Hz, 2 to 4 dB
De-esser	Sibilance	5 to 9 kHz, 3 to 6 dB of reduction on peaks only
Compression	Level consistency	3:1 to 4:1, 3 to 6 dB, medium attack
Second compressor	Catching the loudest moments	Fast, 2 to 3 dB, only on peaks
Additive EQ	Presence and air	Broad lift at 3 kHz and above 10 kHz
Saturation	Density and forwardness	Subtle. It should read as loudness, not distortion

Two gentle compressors in series almost always sound better than one working hard. The first levels the performance, the second catches what got through.

The mastering chain in the order each stage assumes.

Making a vocal sit without making it loud

Carve the beat, not the vocal. A narrow 2 to 4 dB dip around 2.5 kHz in the instrumental opens a slot.
Use parallel compression so the quiet detail comes up without flattening the peaks.
Automate before you compress harder. Fader rides fix problems compression cannot.
Keep the reverb send high-passed above 300 Hz so it adds depth without mud.
Check in mono. If the vocal moves or thins out, the width treatment is wrong.

Recording decisions that save the mix

Get the singer 15 to 20 cm from the mic with a pop filter in between.
Track at 24-bit and peak around -12 dBFS. Headroom costs nothing.
Treat the reflection points behind and beside the singer, not just the wall in front.
Record three usable takes rather than one perfect one. Comping beats repairing.
Kill the noise floor at the source. Restoration always costs something.

How it translates on real systems

Almost nobody hears your record on the system you made it on. The point of every decision above is that it survives the journey.

System	What it exposes	What to check
Phone speaker	No low end at all below roughly 500 Hz	Does the bass line still read from its harmonics?
Earbuds	Exaggerated width and sub, hyped top	Is anything sibilant or fatiguing?
Laptop	Thin midrange-only playback	Is the vocal still intelligible?
Car	Boomy low end and road noise	Does the low end turn to mush at 80 to 120 Hz?
Club system	Everything below 40 Hz, loud	Is the sub mono and controlled?

If it works on a phone and in a car, it works almost everywhere else. Those two are the honest tests.

The mistakes that cost the most

These are not exotic. They are the five things that show up again and again in tracks that almost work.

Judging at a different level each time. Loud always sounds better, so set a monitoring level and leave it.
Solving a balance problem with a plugin. If the fader cannot fix the relationship, a compressor will only make the wrong balance more consistent.
Working only on one system. A mix that only holds together on your monitors is not finished, it is calibrated to your room.
Boosting where you should be cutting. Making something louder to beat a masking problem raises the whole noise floor of the arrangement.
Never bypassing. If you have not compared against the unprocessed version in the last twenty minutes, you have lost your reference point.

Take a break every forty minutes. Ear fatigue does not announce itself, it just quietly moves your idea of what bright means.

Frequently asked questions

What is the most important thing to get right with De-Esser online?

Tonal balance and translation. A track that holds up on a phone, in a car and on monitors is finished. One that only works on your monitors is not.

What loudness should I target?

-14 LUFS integrated with a -1 dBTP ceiling covers almost all streaming. Go quieter for acoustic and classical material, louder only where the genre genuinely demands it.

How much headroom should the mix have?

Three to six decibels, with no limiter on the mix bus. That gives the mastering stage something to work with.

How do I know when it is finished?

When it holds up level-matched against a commercial reference, translates on several systems, keeps its low end in mono, and your changes have stopped improving it.