Guide · Vocals

How to de-ess a vocal (without dulling it)

Sibilance is the piercing 'sss' and 'tsh' energy that sits roughly between 5 and 9 kHz. A little is presence; too much is fatiguing. De-essing turns it down only when it spikes — so the vocal keeps its air the rest of the time.

Updated 2026-07-06

Find the sibilance first

Before you cut anything, find the exact frequency. Sweep a narrow boost through 5–9 kHz while the vocal plays; the spot that makes the 's' sounds screech is your target. Female and brighter voices tend to sit higher (7–9 kHz), deeper voices lower (5–7 kHz).

Two ways to de-ess

A dedicated de-esser is a compressor tuned to the sibilant band — simple and fast. A dynamic EQ band does the same job with more control: you place a bell exactly on the offending frequency, set it to dynamic, and dial a threshold so it only cuts on the harsh syllables.

The dynamic-EQ route wins when the sibilance moves around or when a broad de-esser is dulling the whole top end. You keep the brightness on vowels and only clamp the 's'.

  • Narrow-ish Q, centred on the sibilance you found by sweeping.
  • Dynamic cut, not a static one — flat when there's no 's'.
  • Aim for 2–4 dB of reduction on the peaks; more and it lisps.
  • In stereo mixes, de-ess in mid/side and target the centre where the vocal lives.

Don't over-do it

Over-de-essing turns 'sun' into 'thun' — the classic lisp. If you're pulling more than a few dB, the recording or the mic choice is usually the real problem. Fix the loudest offenders, then leave the character alone.

Free · no license key

De-ess in one click with AURORA

AURORA's Smart Ops includes a de-ess pass, and any band flips to a dynamic cut you can place exactly on the sibilance — Peak or RMS detector, threshold, ratio and range. Free.

Download AURORA free →
FAQ
What frequency is vocal sibilance?

Usually 5–9 kHz. Brighter and higher voices sit toward 7–9 kHz; deeper voices toward 5–7 kHz. Sweep a narrow boost through that range to find the exact spot for your take.

Should I de-ess before or after compression?

Usually after the main compression, because compression brings up sibilance. Many engineers de-ess late in the chain, right before any bright EQ or the final limiter.

Can I de-ess with a normal EQ?

A static EQ cut will dull the vocal on every word. Use a dynamic EQ or de-esser so the cut only happens on the harsh syllables and the vocal keeps its air otherwise.

Get launch news + free drops
Launch news only. No spam, unsubscribe anytime.