Home Vocal Recording and Mixing

How to de-ess vocals in REAPER (with stock plugins)

By StudioRescue Editorial

Published July 23, 2026

Sibilance is the sharp, hissing energy in speech and singing that comes from the “s,” “sh,” “ch,” and “t” sounds. A little is normal and even helps a vocal sound clear. Too much is fatiguing, especially on earbuds and phone speakers, and it tends to jump out even more after you compress the vocal.

The good news is that sibilance is a level problem in a narrow range of frequencies, not a performance problem, and REAPER can reduce it without any paid plugins. The trick is to reduce the esses only when they get loud, instead of dulling the whole vocal. Here is how to find the problem and fix it in REAPER with stock tools.

What sibilance actually is

Sibilance is high-frequency energy that spikes for a fraction of a second on certain consonants. Because it comes and goes with the words, it is a dynamic problem. It is loud on one word and gone on the next.

The frequencies involved sit roughly between 2 kHz and 10 kHz, with the most annoying esses usually landing around 5–8 kHz. Where exactly depends on the singer and the microphone. Lower or deeper voices often peak nearer 5–6 kHz, while brighter or higher voices tend to peak nearer 7–8 kHz. A cheaper, brighter condenser can push the peak around too. Treat those numbers as a place to start listening, not a fixed setting.

Common reasons a home vocal gets too sibilant:

• The microphone. Many affordable condensers have a lifted high end that exaggerates esses.
• Mic technique. Singing straight into the capsule, close up, aims the sibilance right at the mic.
• Compression. Heavy compression on the vocal raises quiet detail, and that includes the esses, so a vocal can get more sibilant after you compress it.

Why a plain EQ cut is the wrong tool

It is tempting to just pull down the high frequencies with an EQ and call it done. The problem is that a normal EQ cut is always on. It removes the same amount whether the singer is on a bright “s” or holding a soft vowel, so it dulls the entire vocal to fix a few moments.

De-essing is different because it only acts when the ess is loud. It listens to the sibilant range, and when the energy there crosses a threshold, it turns that range down for that instant and then lets go. That is why a de-esser keeps the vocal bright and clear between the esses while still taming the harsh spikes.

Find your sibilance frequency first

Guessing at the frequency is how vocals end up sounding like they have a lisp. Locate the problem before you reduce anything.

Three ways to de-ess for free in REAPER

REAPER does not include a plugin called “de-esser,” but three stock approaches all do the job. Pick one. Put the de-essing stage after your compressor, since compression is often what makes the esses jump.

1. ReaComp with a detector filter (easiest)

ReaComp can be told to listen to only the high frequencies before it decides to compress. That turns a normal compressor into a de-esser.

• Add ReaComp to the vocal after your main compressor.
• Find the detector filter controls and raise the high-pass so the compressor mostly hears the sibilant range, roughly your swept frequency and above.
• Turn on the Preview filter option to hear what the compressor is listening to, confirm it is mostly esses, then turn preview back off.
• Use a fast attack and a fairly quick release so it grabs the short ess and lets go.
• Lower the threshold until you see about 2–6 dB of gain reduction on the esses only, and little or no movement on the rest of the words.

2. ReaXcomp on the top band (cleaner)

ReaXcomp is REAPER’s stock multiband compressor. Because it splits the signal into bands, it can turn down only the sibilant band and leave the rest of the vocal untouched, which usually sounds cleaner than the ReaComp method.

• Add ReaXcomp to the vocal.
• Set a band crossover so the top band starts around your sibilance frequency, near 6–7 kHz for many voices.
• Lower the threshold on that top band only, aiming for a few dB of reduction when the esses hit.
• Leave the lower bands doing nothing, so only the ess range is controlled.

3. Manual clip gain or a volume envelope (most control)

You can also de-ess by hand, with no plugin at all. Zoom in on the waveform and the esses are usually easy to spot as short, dense spikes.

• Use clip gain, or a pre-FX volume envelope, to pull each loud ess down by about 2–6 dB.
• This gives you precise control over each word, and it is easy to undo if you go too far.
• It is slower, but for a short vocal or a stubborn line it often beats any plugin.

What de-essing will not do

De-essing is a targeted tool. It helps a lot with esses, but it has limits.

• It will not fix a vocal that is harsh across the board. Constant edge on vowels is upper-mid tone, and it needs EQ, not a de-esser.
• It will not remove distortion or clipping baked into the recording. Damaged highs stay damaged.
• Pushed too hard, it creates a new problem: the vocal starts to sound like it has a lisp, because you have removed the esses the words need to be understood.
• It cannot fully undo a very bright or spitty microphone. It reduces the symptom, but a better capture, or a small change in mic angle, prevents it in the first place.

If it still sounds off

Quick troubleshooting order when de-essing is not landing:

1. It sounds lispy. You reduced too much. Raise the threshold so only the loudest esses are caught, and use less reduction.
2. It is dulling the whole vocal. Your target frequency is too low or the band is too wide. Move it up and narrow it toward the true ess peak.
3. The esses still poke through. The de-esser may be before the compressor. Move it after, so it catches the esses the compressor pushed up.
4. Only one or two words are bad. Skip the plugin and fix those with clip gain. It is faster and cleaner than de-essing the whole take.