sonible Smart Deess Instinctual Sibilance and Plosive Removal User Guide
- June 16, 2024
- sonible
Table of Contents
sonible Smart Deess Instinctual Sibilance and Plosive Removal
Welcome to smart:deess
- smart:deess is an AI-powered de-esser, that combines plosive reduction with advanced spectral de-essing.
- The finely nuanced composition of the human voice needs handling on a granular level. While conventional de-users just reduce volume, smart:deess goes far deeper for a more natural, better-balanced sound. The plug-in unites spectral processing and a real-time phoneme detection network to automatically improve sibilants and plosives with supreme precision, making it essential for the two most common problems with vocal recordings.
- Thanks to its AI-powered processing, the plugin only needs to be loaded on a track to start fixing a vocal recording in a way that’s tailored precisely to your source audio’s characteristics.
System requirements
CPU
- Intel Core i5
- Apple M1
RAM
- 4GB
Operating systems
- Windows 10+ (64-bit)
- Mac OS 10.14+
Graphics
- OpenGL Version 3.2+
You will need admin privileges to successfully install the smart:deess plug- in.
Mac OSX
- To start the installation process, please open the disk image sonible_smartdeess_osx_x.x.x.dmg. This will mount the image and open a finder window showing the content of the installation package.
- To install smart: dress on your system, run the installation file smartdeess. pkg.
- The installer will now guide you through the necessary steps to install smart:deess on your computer. smart: dress will automatically be installed in the default locations for audio plug-ins.
- Default folders:
Audio Unit
- /Library/Audio/Plug-Ins/Components/
VST
- /Library/Audio/Plug-Ins/VST/
VST3
- /Library/Audio/Plug-Ins/VST3/
AAX
- /Library/Application Support/Avid/Audio/Plug-Ins/
Windows
To start the installation process, extract the downloaded zip file
sonible_smartdeess_win_x.x.x.zip onto your hard disk and run the installer.
The installer will now guide you through the necessary steps to install
smart:deess on your computer.
smart: dress will automatically be installed in the default locations for
audio plug-ins.
Default folders:
VST3
- C:\Program Files\Common Files\VST3\
VST
- C:\Program Files\Common Files\VST\
AAX
- C:\Program Files\Common Files\Avid\Audio\Plug-Ins
Authorization
Licensing system
You can select between two licensing systems: machine-based or iLok (USB
dongle).
By creating a user account on www.sonible.com and
registering your products – if they are not already visible in your Dashboard,
you can manage your plug-in activations.
Machine-based
- Each license key allows you to install smart:deess on two computers with unique system IDs. These system IDs are computed during license activation.
- The same license can be used by multiple users, but each user has to individually unlock the full version of smart: dress under their account.
- In case a system ID is changed (e.g. replacement of the hard drive), you can revoke/activate the plug-in next to the respective system ID in the Dashboard of your sensible user account.
iLok
If you want to transfer one activation to your iLok, just make sure the plug-
in is registered in your sensible user account. Click on the button „transfer
to iLok“ next to the plug-in in your Dashboard and follow the instructions.
Note: 1st gen iLok dongles and the iLok Cloud are currently not
supported.
Unlocking
If you purchased a license for Smart: dress online, you receive your license
key via email.
Machine-based unlocking
When opening Smart: dress for the first time, a notification window will be
displayed asking you to unlock Smart: dress with a valid license key.
Please make sure that your computer is connected to the internet before
starting the registration process.
Enter your license key and click „register.“ The plug-in will now communicate
with our server to check if the license is valid. If it is – enjoy! 🙂
iLok
If you transferred your license to an iLok, simply attach the iLok to your
computer. The plug-in will then be automatically registered – enjoy!
If you don’t receive the email within minutes, please check your junk folder
first before contacting our support
(support@sonible.com).
Trial version
To run smart: dress in demo mode, simply click ”try” and you will then be able
to use smart:deess for a couple of days without any limitations. (Please refer
to our website to find out more about the current demo period of smart:deess)
When the demo period expires, you will need to purchase a full license to
continue using the plug-in.
Internet connection requirements
sensible plug-ins only need an internet connection during the trial period and
for initial license activation. During the trial period, the plug-in needs to
go online every time it is used. Once the license of your plug-in has
successfully been activated, an internet connection is no longer needed.
User interface
All parameters or visualizations on the user interface related to sibilants are marked in green, while elements related to plosives are marked in blue.
Getting started with smart:deess
-
Analyzing the Input Signal
Setting up smart:deess is a push of a button. When opening the plug-in, smart:deess asks you to playback some audio to learn a short voiceprint from your signal. This analysis allows smart:deess to adapt its processing to the characteristics of your speaker or singer.
INFO: What is a voiceprint and why does smart:deess need to learn that for each speaker? smart:deess analyzes the signal’s actual content – not just the level of one frequency band – to make decisions. A ‘voiceprint’ of the input voice signal is like a fingerprint of the vocal characteristics of a singer or speaker. This vocal fingerprint – we call it voiceprint – is analyzed by smart:deess and the internal processing is adapted to match the observed characteristics. -
Processing Sibilants and Plosives
Once a signal has been analyzed and has been learned, smart:deess is ready for processing. Sibilants are highlighted in green and plosives are highlighted in blue in the input signal. Your job now is to review the analysis and refine the results (see next section), spending time on getting the sound you want, not correcting the sound you don’t. -
Reanalyzing Voice
When working with a new speaker or singer, you should always reanalyze the signal to make sure that the processing of smart:deess adapts accordingly.
Working with Sibilants
smart:deess identifies specific sibilants (e.g. S, Z, Sh, Ch) in real-time,
determining the exact start and end point of these phonemes. So, the entire
sibilant is processed, rather than just the loudest part, making the
processing sound very natural.
While the processing itself will be natural, different users may have
different ideas about how an “S” should sound like. Also, the character of
sibilants always depends on the context in which they are used (e.g. lead
vocal, backing vocal, podcast, etc.). To accommodate these user- and
application-dependent goals, smart:deess offers several options for fine-
tuning the processing of sibilants.
-
Suppression
The (Sibilant) Suppression parameter controls the overall target level of sibilants. The higher the suppression, the more annoying or loud sibilants are tamed.
After learning a voiceprint, smart:deess will always suggest an initial suppression strength of 50. This setting typically leads to balanced sibilants that are well embedded into the overall level of adjacent phonemes.
Smaller suppression values help to preserve the liveliness of sibilants while larger values guarantee a very homogeneous level of all sibilants. -
Colour
The Color control tunes the sibilant character (soft, balanced or sharp) to get the desired sibilant sound. This can help to tame harsh sibilants or revive dull and lisping sibilants thanks to spectral processing.- soft: The soft colour leads to soft, unobtrusive sibilants. This can be great for smooth vocals or background vocals.
- balanced: The balanced colour tries to preserve the natural colour of an observed sibilant. Only problematic resonances or spectral imbalances are corrected.
- sharp: The sharp colour allows to creation of very present and clear sibilants that stick out without smashing the overall phoneme balance.
-
Shaping
The shaping parameter controls the amount of spectral shaping towards the chosen target colour.
After learning a voiceprint, smart:deess will always suggest an initial shaping strength of 100. This setting will typically lead to natural-sounding sibilants that are moderately shaped towards the chosen target.
Higher shaping values may significantly change the character of all sibilants and lead to very homogeneous sibilant sounds.
Low values for the shaping parameter leave the shape of the sibilants more or less untouched.
HINT: When the Shaping parameter is set to zero, no spectral processing is applied. This can help if you only want to apply a broadband gain reduction to all sibilants.
Working with Plosives
smart:deess identifies specific plosives (e.g. P, T, K) in real time, determining the exact start and end point of these phonemes.
-
Controlling the level of plosives
The (Plosive) Suppression parameter controls the overall target level of plosives. The higher the suppression, the more disturbing and loud plosives are tamed.
After learning a voiceprint, smart:deess will always suggest an initial suppression strength of 50. This setting will typically lead to well-balanced plosives that are embedded into the overall level of neighbouring phonemes.
Smaller suppression values help to preserve the liveliness of sibilants while larger values guarantee a very homogeneous level of all sibilants. -
Shaping
The shaping parameter controls the amount of spectral processing for plosives. Other than for sibilants, smart:deess does not use a target colour for plosives, but simply balances out spectral problems like harsh resonances or boomy low ends of plosives.
After learning a voiceprint, smart:deess will always suggest an initial shaping strength of 100. This setting will typically lead to well-balanced plosives with a natural articulation.
Higher shaping values can help to create very smooth sounding plosives for a subtle and homogeneous sound.
Low values for the shaping parameter leave the shape of the plosives more or less untouched.
Processing Mode and Fine-Tuning
Processing Mode
In Full Range mode, the whole spectrum is processed. In Split Band mode, only
the chosen frequency ranges are spectrally processed.
Full Range: In most situations, it is recommended to use Full Range mode
as this mode typically generates the most natural results. In Full Range mode,
the whole spectrum is processed, meaning that the broadband gain reduction
(reduction of the level of phonemes) is combined with the spectral processing
(enhancing the shape of phonemes) across the whole frequency range.
When setting the two Shaping values for sibilants and plosives to zero,
processing in full range mode is equivalent to simply turning down the level
of sibilants and plosives.
Split Band : In Split Band mode, the processing ranges for sibilants and plosives can be limited to specific frequency ranges. For sibilants, an upper and lower threshold can be set; for plosives, only an upper threshold can be set as the low end should always be processed. Working in Split Band mode can help to leave certain frequency ranges completely untouched. This can help if you are processing very dense signals, where you only want to process the most characteristic frequencies for sibilants or plosive. Outside the chosen frequency ranges, no gain reduction is applied to the respective phonemes. You can also monitor this frequency-limited processing in the spectrogram. The chosen frequency ranges for sibilants and plosives can be soloed by clicking the corresponding headphone icon.
Diff. Signal
Listen to the difference between the input and the output signal. When the Diff. Signal mode is active, the signal history will highlight the delta signal.
Presets & States
Settings
CONTACT
- sensible GmbH
- Haydngasse 10/1
- 8010 Graz
- Austria
- contact@sonible.com.
- www.sonible.com.
All specifications are subject to change without notice. ©2023, sonible GmbH. All rights reserved. Engineered & designed by Sonible in Austria.