Taming harsh vocals with smart:deess

smart:deess revolutionizes de-essing and plosive removal. By tackling both problems with supreme precision, this plug-in is an essential in your workflow. We show you how easy it is to fix vocal recordings with smart:deess.

For more detailed information on all the features of smart:deess download the latest manual.

Step-by-step instructions

smart:deess is the world’s first AI de-esser. Improve your vocal tracks with automatic phoneme detection and spectral de-essing.

Conventional de-essers just reduce volume, triggered by the level of certain frequency bands.

smart:deess uses a neural network to identify the start and end points of phonemes and applies distinct spectral processing to each of them – delivering the most natural and highly customizable results.

To tame the harsh “ess” sounds on any recording, you need to start the audio playback and let smart:deess analyze your vocal track. After learning, smart:deess will adapt its processing to the audio material.

Increase the Sibilant Suppression to further reduce the level of your sibilants.

Change the Target Color and Shaping parameter to further fine tune your sibilants’ and plosives’ sound.

Monitor your changes in the Gain reduction plot and spectrogram.

When pushing conventional de-essers, voices quickly start to sound lispy. smart:deess sounds natural even when doing some heavy lifting.