Shusssh

Start with Sensitivity while the vocal plays against the music. Raise it until the distracting consonants are consistently recognised, then adjust Amount for how much treatment the phrase needs. These controls solve different problems: reacting more often is not the same as reducing an event more strongly.

Leave Focus at AUTO initially. Move it lower or higher only when the automatic emphasis misses the offending region. Focus biases a moving detector; it is not a fixed frequency dial or an eight-kilohertz preset for every singer.

For a bright lead, compare Shusssh before and after the stage that makes the consonants conspicuous. Choose the position that lets you use the least treatment. Match with Output and listen for the beginning and end of each word, not just the disappearance of the sharp sound.

For stacked backing vocals, try one instance on the stereo group and compare it with lighter processing on the individual takes. A useful result leaves the stack clear without making every singer sound identically softened.

A dedicated vocal send is another useful place to try Shusssh. Treat the copy feeding a bright reverb or long delay so an isolated consonant does not dominate the tail. Keep the dry lead on its own path and judge the return in the arrangement.

The engine combines a spectral mask with limited wideband action and gain compensation. Use the UI Listen command to inspect the treatment, then return to normal playback to make the musical decision.