Updated October 2026 · Vocal recording fixes
De-essing reduces the harsh hiss of S, SH and T sounds (sibilance). Find the frequency where the hiss is strongest, set a compressor-style de-esser to reduce only when that band gets loud, then listen in context. Sibilance often sits roughly between 4 and 9 kHz, but it depends on the voice and the mic.
Solo the vocal and loop a phrase with several S sounds. Sweep a narrow EQ boost across the top of the voice until the hiss jumps out. That frequency is where the de-esser should work.
Bypass the de-esser and compare. If the vocal sounds lisped or dull, you are reducing too much or the wrong frequency. Check on phone speakers too, where sibilance is easy to miss.
A de-esser cannot fix a bad recording. If the mic is too close or the singer is very bright, reduce the top end a little more gently and record with the mic placed off-axis next time.
Spot clipping, noise, rumble and level problems. Runs in your browser, nothing is uploaded.
Often roughly 4 to 9 kHz, but find yours by listening. The exact number depends on the voice and the microphone.
Yes. Over-processed sibilance sounds dull and lisped. Reduce the amount until the words still sound clear.
Any de-esser works. What matters is setting the frequency and threshold by ear.
