How to Recognize This Problem
- S and SH sounds in vocals are significantly louder and harsher than surrounding consonants
- The vocal sounds "spitty" or like the singer is hissing at the microphone
- Compression makes the problem worse — sibilant peaks become even more prominent
- High-frequency limiting or clipping on the mix bus triggers audibly on vocal sibilance
- The vocal sounds fine solo but the sibilance becomes painful when the full mix is playing
Why This Happens
Bright Condenser Microphones
Large-diaphragm condensers with a presence peak at 5-10kHz exaggerate sibilance by design. This brightness is flattering on some voices but creates unbearable sibilance on naturally sibilant singers.
Close Microphone Distance and On-Axis Positioning
Singing directly into the microphone capsule on-axis at close range concentrates the high-frequency energy of sibilant sounds. Even a slight off-axis angle can reduce sibilance dramatically.
Compression Amplifying Sibilant Peaks
Compressors reduce the dynamic range of the vocal body but cannot catch the ultra-fast transients of sibilance (2-5ms attack). The result is that sibilant peaks now sit 6-10dB above the compressed vocal body, making them disproportionately loud.
Cumulative High-Frequency Processing
Presence boosts, air EQ, exciter plugins, and bright reverbs all add energy in the sibilance range. What starts as subtle sibilance in the recording becomes piercing after passing through a chain of high-frequency-enhancing processors.
How to Fix It
Apply De-Essing Before Compression
Insert a de-esser before your main vocal compressor. Set the frequency to the problem range (usually 5-8kHz) and adjust the threshold so it catches only the sibilant peaks, reducing them by 3-6dB. Listen carefully — over-de-essing creates a lisp.
Use Split-Band or Wideband De-Essing Strategically
Wideband de-essers reduce the entire signal when sibilance is detected, which sounds more natural. Split-band de-essers only reduce the high frequencies, which is more transparent but can sound unnatural if pushed too hard. Try both and choose what sounds best for the voice.
Manual Clip Gain Reduction on Worst Offenders
Zoom into the vocal waveform and reduce the clip gain of the most egregious sibilant peaks by 3-6dB before they hit any plugins. This is time-consuming but gives the most transparent results because you are editing the source level, not processing.
Adjust Microphone Technique for Future Sessions
Position the mic slightly off-axis (15-30 degrees) so sibilant sounds hit the edge of the capsule rather than dead center. Alternatively, use a dynamic mic for sibilant singers — the SM7B tames sibilance naturally compared to bright condensers.
Apply a Dynamic EQ as a Precision De-Esser
A dynamic EQ band at 6-8kHz with a narrow Q acts as a surgical de-esser. Set the threshold so it only engages on sibilant peaks. Unlike a broadband de-esser, this targets only the exact problematic frequency and leaves surrounding harmonics untouched.
What RoastYourMix Measures
RoastYourMix does not detect sibilance or isolate the vocal. It measures the high mids (2-8 kHz) and highs (8-20 kHz) of your whole mix in dB against the typical range of thousands of real uploads, so a mix that is bright overall shows up there, but harsh S sounds on their own are a job for your ears and a de-esser.
The quick answers
Can I completely remove sibilance?
You should not try to remove it completely — sibilance carries essential consonant information that makes lyrics intelligible. The goal is to bring sibilant peaks down to a level consistent with the rest of the vocal, not to eliminate them. A good de-essing job is one you cannot hear working.
Why does my de-esser make the vocal sound lispy?
The threshold is set too low, causing the de-esser to engage on non-sibilant content and reduce too much high-frequency energy. Raise the threshold until it only catches the most prominent sibilant peaks, and reduce the amount of gain reduction to 3-6dB maximum.
Is sibilance worse with certain singers?
Yes. Sibilance varies significantly by voice, language, and singing style. Female vocals tend to have sibilance at higher frequencies (7-10kHz) while male vocals concentrate lower (4-7kHz). Some singers are naturally more sibilant due to dental structure and articulation habits.