To use a channel strip as a de-esser you need to key the dynamics with an external signal to only compress the frequency which contains the sibilance. You will need two units, each being sent the same audio signal. You use one unit to EQ the signal to isolate the sibilance frequencies. This can be achieved by using a tight Q HMF gain at the sibilance frequency, combined with filtering or EQ reduction of all other frequencies. Feed the output of this unit into the dynamics key of the second unit. Make sure dynamics key is pressed on the second unit. The output from the second unit should now be de-essed.
Sibilance is defined as the presence of strongly emphasized s, sh, ch, z, j sounds in speech. These are created by air moving through the vocal tract and being constricted by the position of the tongue and lips. The spectrum of sibilants is in the high frequency range (5-10 kHz). A De-esser compresses this narrow bandwidth to reduce the audible effect of sibilance.
