If your S sounds cut through the mix like a hiss or your T’s feel spiky in headphones, you have sibilance, and a de-esser is the tool that tames it. It is a common finishing touch on professional voice chains, but not every streamer needs one. Knowing what it does helps you decide whether to add it to your OBS filter stack.
What sibilance is
Sibilance is the harsh, high-frequency energy in consonants like S, SH, T, and Z, usually concentrated between about 5kHz and 9kHz. Some voices, mics, and close-mic techniques exaggerate it. Bright condenser mics and boosted high-frequency EQ can make it worse, turning otherwise clean audio into something fatiguing to listen to for hours.
How a de-esser works
A de-esser is a frequency-specific compressor. It listens for energy in the sibilant range and briefly reduces gain only when those harsh sounds spike, leaving the rest of your voice untouched. Unlike a static EQ cut, it only acts when sibilance is present, so it does not dull your whole voice.
De-esser vs other tools
| Tool | What it fixes | When to use |
|---|---|---|
| De-esser | Harsh S/SH/T spikes | Sibilant voice or bright mic |
| EQ (static cut) | Constant tonal issues | Broad tone shaping |
| Compressor | Overall level consistency | Evening out volume |
| Noise gate | Background noise between words | Noisy rooms |
Do you actually need one?
You need a de-esser if you can clearly hear harsh, whistling S sounds in your recordings, especially after adding compression or high-end EQ, which both amplify sibilance. You probably do not need one if your voice already sounds natural and smooth. It is a corrective tool, not a mandatory part of every chain, adding one to a voice that is not sibilant just risks a lisping, dull sound.
Setting it up in OBS
OBS can host VST plugins, so you can add a free de-esser VST as a filter on your mic. Place it after your compressor in the chain, since compression brings out sibilance. Set the target frequency around 6–8kHz, then raise the amount just until the harshness eases, stopping before your voice starts to lisp. Less is more.
Dialing in a de-esser correctly
A de-esser has two main controls: the target frequency and the threshold or amount. First find the frequency where your sibilance lives, usually somewhere between 5kHz and 9kHz, by sweeping until the harsh S sounds are most affected. Then raise the reduction amount slowly while speaking sentences full of S and SH sounds, stopping the instant the harshness eases. Push it too far and your S sounds vanish, leaving a lisping, unnatural voice.
Split-band versus wideband
Better de-essers work in split-band mode, compressing only the sibilant frequencies and leaving the rest of your voice untouched, which sounds far more transparent than a wideband version that ducks your whole voice on every S. If your plugin offers a mode switch, split-band is almost always the cleaner choice for voice.
Prevention beats correction
Some sibilance is created upstream. Angling the mic slightly off-axis from your mouth (so you are not talking straight into the capsule) reduces harsh consonants at the source, as does easing back on high-frequency EQ boosts. Fix what you can with technique first, then let a light de-esser catch the remainder.
FAQ
Where does a de-esser go in my filter order?
Typically after noise suppression and compression but before final output limiting. Compression accentuates sibilance, so de-essing afterward catches what it brings out.
Can I fix sibilance with EQ instead?
A static EQ cut around 6–7kHz helps but dulls your voice constantly. A de-esser only acts on the spikes, preserving clarity the rest of the time.
Should the de-esser go before or after compression?
After. Compression raises quieter parts of your voice and, in doing so, accentuates sibilance, so placing the de-esser after your compressor lets it catch the harsh S and T sounds the compressor brought out. A typical chain is noise suppression, then compression, then de-esser, then any final limiting. Putting the de-esser first means it works on sibilance the compressor will re-amplify afterward, defeating its purpose.
Bottom line
A de-esser is a targeted tool for harsh S and T sounds. Add one only if you actually hear sibilance, place it after compression, and dial in the minimum amount that smooths the harshness without making you sound like you have a lisp.
