An “s” sound can feel disproportionately sharp even when the rest of a voice is comfortably recorded. A de-esser is intended to control that kind of problem selectively; Adobe’s Audition effects reference describes its DeEsser in that role. Applying it heavily to an entire recording can also make speech dull or unnatural.
Find the actual moments
Listen at a moderate level and mark two or three troublesome syllables. Include neighbouring words in each selection; an isolated consonant gives little sense of how the voice should sound in context. Compare those moments with a naturally bright vowel or a soft breath that you want to preserve.
Use the effect’s audition or monitoring option if it provides one, following the tool’s documentation. Start conservatively and change one control at a time. A setting copied from another speaker may target the wrong part of this voice.
Judge the phrase after processing
Alternate between the original and processed phrase at comparable playback levels. Ask whether the sharpness is less distracting and whether the words remain clear. If an “s” starts sounding lisped or disappears, reduce the processing and listen again.
Keep an unprocessed version and record the settings used. Some problems are better handled by a small local level adjustment or a new recording than by stronger global processing. The aim is intelligible, natural speech in this episode’s context, not eliminating every high-frequency sound. A quiet breath and an overemphasised consonant can occupy neighbouring frequencies while doing very different work for the listener.
AI-generated editorial illustration; not a documentary photograph.

