A Guide to De Esser

By Daniel Cooper 10 August 2025 3 min read

Filed under

Add Viyolens.COM to your preferred sources in Google

A de-esser is a dynamics processor designed to reduce sibilance — the harsh, hissing quality that appears on the letters "s", "sh" and "ch" in vocal recordings and voiceovers. Sibilance is particularly noticeable in close-mic setups, brighter condenser microphones and digital recordings where high-frequency transients are captured cleanly. Rather than applying broad EQ cuts that affect the whole signal, the processor responds only when sibilant energy crosses a set threshold, leaving the rest of the vocal untouched.

What a De-Esser Does

The processor monitors a specific frequency band — typically between 4 kHz and 12 kHz — and applies gain reduction only when the energy in that band crosses a threshold. In wideband mode, the gain reduction affects the entire audio signal whenever a sibilant is detected. In split or frequency-selective mode, the reduction is applied only to the targeted frequency band, leaving the rest of the signal at full level. Split mode is generally preferred for vocals because it causes less audible pumping on the overall signal.

Sibilance and Microphone Choice

Certain microphones emphasise the upper midrange and high frequencies, which makes sibilance more pronounced in recordings. Large-diaphragm condenser microphones — the standard choice for studio vocals — often have a presence peak in the 8–12 kHz range that can exaggerate harsh consonants. Recording position also matters: close-miking and on-axis placement increase sibilant capture. A de-esser can correct sibilance introduced by microphone characteristics or room acoustics, though it works best when the source recording is as controlled as possible.

Where It Sits in the Signal Chain

In mixing, the processor typically sits after equalisation and before or after compression on the vocal channel. Placing it before compression avoids the compressor reacting to sibilant peaks and creating pumping artefacts. Some engineers prefer placing it after compression to catch any sibilance that compression emphasises. For an overview of the available hardware and software options, the buyer's guide covers the main choices at each price point.

Frequently Asked Questions

What is a de-esser used for?

It targets harsh sibilant consonants — s, sh and ch sounds — in vocal recordings, applying gain reduction only when sibilant energy crosses a threshold. This removes the hiss without applying broad equalisation that would dull the whole vocal.

What frequency does sibilance occur at?

Sibilant energy typically falls between 4 kHz and 12 kHz, with the precise peak varying by voice and microphone. Most processors let you sweep the detection frequency to find the problem area on a specific vocal rather than applying a fixed setting.

What is the difference between wideband and split mode?

Wideband mode applies gain reduction to the entire signal when a sibilant is detected. Split mode applies reduction only to the targeted frequency band, leaving the rest of the signal unaffected. Split mode produces a more transparent result and is preferred for most vocal work.

Does sibilance removal affect vocal quality?

When set correctly, the effect is inaudible — listeners hear a cleaner vocal without noticing any processing. Over-processing creates a lispy quality and removes the vocal's natural brightness. Always compare the processed result with the unprocessed signal to check that the correction is not excessive.

Can I use equalisation instead?

A static EQ cut at the sibilant frequency will reduce harshness but also reduces vocal brightness continuously, even when sibilance is not a problem. Dynamic processing targets only the moments when sibilance occurs, preserving the vocal's natural tone the rest of the time.