The clarity of your dialogue or vocals can easily make or break a project.
If you’re working on producing a song or you’ve been in the field shooting for your film, and between your cheap hire mics and your buck-toothed lead your edit sounds like the inside of Tweety-bird’s cage instead of the atoms of your moody arthouse romance, then there are some very basic things you should know if you want to become a pro.
A De-esser is a compression plugin that specifically targets unwanted noise in audio files of recorded voices: specifically the ‘ess’ noise they are named after.
This ‘sibilance’ is the hissing noise that registers on higher frequencies and is loudly picked up on microphones from human speech, such as on words containing ‘ess’ or ‘eff’ sounds.
De-Essers are a fairly standard piece of kit for even the most basic sound edits, and are useful for podcasts, voiceovers & ADR, dialogue editing and song vocals. They’re essentially the combination of using compression and EQ to exclusively cut sibilance frequencies from your dialogue mix.
As such, pretty much every audio workstation fields its own De-Esser compression software- but many of them don’t have the versatility a seasoned or burgeoning pro, such as yourself, expects.
Handily, the world is just chocked-full of expertly-made characterful audio plugins competing for your dollars. We’re taking a look at what we consider to be the very best in De-Esser plugins available to you:
- 1 FabFilter ProDS
- 2 Oeksound Soothe2
- 3 Softube Weiss Deess
- 4 Eiosis E2Deesser
- 5 Waves Sibilance
- 6 Buyer’s Guide
- 7 Frequently Asked Questions
One of the best audio software developers in the business, and you can really tell on this bit of kit: this is possibly one of the most scalable and versatile plugins out there. Every single imaginable aspect of De-Essing and De-Essing-related activities are controllable by you.
Functions are divided between ‘Allround’ and ‘Single Voice’: as might be expected, the first can be used to soften and remove harsh frequencies from all manner of sounds or instruments (think hi-hats, or brass instruments etc.), while the latter is supremely focused on removing jarring sibilance from vocals.
FabFilter algorithms can also be unleashed on a clip in either Split-Band or Wide-Band processing, depending on whether you want to affect the entire frequency spectrum or high frequencies over a threshold point that you have chosen.
An underrated feature that is sorely lacking from many De-Esser plugins is a visual feedback of your adjustments in real time; leave it to FabFilter to break the mold. Well recommended.
- Intelligent systems – Algorithms purely select sibilance to retain your audio
- Jack of all trades – Versatile to the most minute details
- Feedback – See how your changes are affecting your mix
- Pricey for a deesser – and plenty of other choices out there
Soothe is a far more subtle software plugin, or ‘dynamic resonance suppressor’ as it is advertised. You place individual bands at the offending frequency and then adjust the depth, sharpness and selectivity controls until you get the sound you’re after.
Soothe is a bit of a sly weapon for any producer who wants to improve their mixes- it’s essentially an auxiliary noise reduction device.
With the upgrade from the original version comes the addition of many new features: Advanced editing powers give you control over the entire hearable spectrum rather than just the standard highs.
You can also apply attack and release to various elements of your adjustments to smooth out the action.
This is an advanced plugin that actually cares more about minimalism- the ethos here appears to be more geared to the qualities of the sounds present in the mix, and not what another compressor can add. Incredibly powerful for mixing vocals.
- More than just a Deesser – Entire frequency spectrum available to you
- Definite upgrade – Cleaner and improved from previous versions
- Smart – Detects and eliminates jarring frequencies
- Pricey if you just want a Deesser – Cheaper alternatives available
With a trial available, there’s every reason to give this one a try. Here is a good clean UI with wonderful functional versatility for whatever you wish to apply it for.
This software is actually based upon a very well-respected piece of hardware: the Weiss DS1-MK3 De-Esser, and here in software form the makers have sought to insert some extra innovation.
Starting of course with that gorgeous user interface, everything has been available for you to play with and adjust. Included in this flexibility is control over attack, release, knee and a filter/threshold system.
The jump to software also means that Weiss Deess is compatible with Windows 10 and higher, and macOS 10.13 and higher, and in VST 2/3, AU and AAX formats. If brand recognition is important to you then this venerated developer is the holy grail.
- Respected and well-known name – market satisfaction behind you
- Free trial – Try before you buy
- Strength in versatility – One of the most flexible De-Essers out there
- Expensive – With brand recognisability comes a higher price tag
Eiosis doesn’t look for frills when it comes to audio software design- this De-esser does the job, requires minimum setup, and still sounds great. This version actually includes a waveform display to make adjusting your setup to hack out that sibilance an absolute breeze.
In the spirit of minimalism, much of the Eiosis works for itself. Playing with the Focus control allows you to do just that- focus- on the frequencies which the Eiosis has already selected as being problematic.
The Shaping controls allow you to soften/sharpen the curve of the de-esser, as represented in the visual feedback. A softer curve means the Eiosis applies a wider curve to the de-esser in your audio, which when processed, sounds far less jarring.
Sharpness gives a tighter curve and will allow you to be far more precise. This is an excellent and simple piece of kit that appeals to sound techs of all abilities.
- Simple to work – and does a lot of the work for you
- Simple to setup – be on you way in minutes
- Simple to understand – visual representation display all your adjustments
- Bare bones – Not as many features as other De-essers out there
This is a de-esser that can be really utilized for non-vocal sibilance. Rather than affecting narrow frequency bands, Waves uses spectral filters to route out undesirable sibilance and separates it from the desirable vocals before it applies processing- similar to noise reduction.
Hopefully, the result of all this fancy tech is that your vocal signal is good and clean and ready for the next step of your mix.
Visualize the reduction as it is applied using the waveform display, to adeptly corner the sibilance in your track; then use threshold setting, wide and split band detection, and detection control to minimize the damage to your desired vocals.
The tool is simple to understand but hard to master like all good creative software, and still provides engineers with everything needed to make effective changes. A look-ahead algorithm allows it to prepare for incoming signals.
The surprising amount of depth on display here makes this an obvious choice for those really keen on delving into their craft and want a piece of software that will grow with them.
- Versatile – Not just good for vocal sibilance
- Simple design & layout – Not going to drown the first-timer in technical jargon
- Incredible amount of depth – Still a must-have tool for the proper techie
- Older design – More modern alternatives on the market
Understanding the features and terminology involved will help you make the best choice when selecting your Deesser. Don’t worry, it’s not all jargon- everything mentioned below will have an impact on how this effect is applied to your audio:
A de-esser is just a specialized compressor after all, so this fairly basic setting will be second nature to anyone who has ever opened a DAW (Digital Audio Workstation).
The Threshold controls the point at which compression is applied to an audio sample: this means that once the amplitude of the audio signal crosses a certain level, the de-esser will begin processing the other settings that you have applied.
This amount of de-essing that takes place depends on the Range value, which is applied once your amplitude has crossed the Threshold you have set.
A lower value means that the algorithm is less aggressive in applying de-essing, and therefore a higher value will make the algorithm more aggressive.
High aggression in this sense allows you to be more discerning and accurate with your applications of de-essing, although this can sometimes create jarring quality changes in your audio. A smoother, less aggressive application softens the blow and is less discernible to the ear.
The de-esser feature ‘predicts’ incoming sibilance by reading the audio ahead of playback and adjusting softly, helping you to avoid over-processing your audio. In styles that feature rapid vocals (such as rapping) or other high-sibilance activities, this can be an extremely useful piece of tech.
Lookahead can cause extra pressure on your CPU, resulting in lag- which is a big no-no for recording, as anyone who has lost hours synching their clips to the track can tell you. Still with a high-powered computer, this is an extremely useful feature.
Wide Vs. Split-band
A Wide-band process is applied to the entire frequency spectrum of your audio clip, while Split-band only affects frequencies after a threshold determined by yourself.
What works best for what performance is entirely dependent on the audio itself, and a certain amount of experimentation is required to find the best results.
If your de-esser allows you to isolate in playback only the sibilance, or only what the processor is affecting, then great: you’ll be able to determine which function is getting the best result for you as you make adjustments.
Frequently Asked Questions
What Is A De-Esser Plugin?
De-essers are a type of compressor that reduces certain unwanted high frequencies in a vocal performance or typical speech; typically those caused by harsh hisses from the pronunciation of sounds like ‘eff’ and ‘ess’.
These noises are known as ‘sibilance’. A De-Esser is a common software plugin for a DAW (Digital Audio Workstation) used in either music or sound editing, and can also be applied to sounds which produce similar unwanted harsh sounds in high frequencies, such as cymbals or wind instruments.
How Does A De-Esser work?
Applying a compressor (in the form of a De-Esser) directly to those frequencies which cause sibilance in an audio clip makes the processed sound softer, and more uniform with the rest of the vocal’s waveform volume.
This is much more pleasing to the ear, and having clear, professional sounding vocals is absolutely vital whatever your medium is. You can often see this effect represented visually by your De-Esser software as you make your adjustments in the form of a spectral EQ graphic.
How Do You Use A De-Esser?
De-Esser plugins come with many different functional tools depending on which software plugin you are using. All of these functions can be used for the purpose of reducing sibilance in a vocal or other performance.
Some functions are standard to all De-Essers:
- Threshold: This controls the point at which compression is applied to your audio: once the amplitude of the audio signal crosses a certain level, the processing settings you have selected will be applied.
- Range: The amount of de-essing which is applied once your amplitude has crossed the Threshold. Can be adjusted for smoothness or precision.
- Lookahead: ‘Predicts’ incoming sibilance by reading the audio ahead of playback by a few milliseconds. Is useful for smoothing out attacks of rapid vocal performances, such as in Hip-Hop.
- Wide-band/Split-band: A Wide-band process is applied to the entire frequency spectrum, while Split-band only affects frequencies after a certain threshold.
- Tip: Toggle the De-Esser on and off as you make adjustments so that you hear the raw clip and your De-Esser processed product so that you have a reference for the changes that you are making.
What Frequency Should I Set a Deesser To?
The frequency of the compression should be adjusted to the area in which sibilance is active. In patterns of speech or vocal performances, this could be anywhere between 4kHz and 7kHz.
Many good plugins will be able to isolate sibilance frequencies, both audibly and even represented visually.
Being discerning and careful with de-essers can minimize quality loss to desirable frequencies. The threshold can prevent compression before an amplitude set by yourself, for example. Range can be used to make precise dents in sibilance with the aid of the spectral EQ.
Is A Deesser Free?
Almost all DAWs include De-Essers as part of their standard in-built package; however these never have the depth and quality of specialized, professionally-made plugins.
Features and quality of output from different software varies wildly between plugins, and each producer or sound editor develops their own taste depending on their ear.
A well-respected plugin can cost quite a bit more than you expect from what is essentially a small part of vocal production. If you are involved in music production on a budget, you may decide that your money is best spent on plugins from other areas of mixing.
How Can I Record My Voice Without Sibilance?
There are a number of things that you can try to reduce sibilance in your audio signal in the first place; after all prevention is much better than fixing in post.
Distance yourself appropriately from the microphone: Many microphones have a sweet spot- there are many types of microphone that record in many different ways, often from different directions.
You’re looking to capture the most from your voice, while also minimizing unwanted peeking from plosive syllables.
Tilt the microphone marginally off-axis: As mentioned, many microphones record differently depending on your position relating to the head. Deliberately recording elsewhere in the microphone’s range can have a big impact on your sound and is a legitimate recording technique.
Fix with EQ: Essentially related to what a De-Esser does anyway, EQing is a huge part of vocal mixing and reducing or manipulating those higher frequencies over the course of multiple EQs is par for the course for even the most basic vocal recordings.
Ride the levels/Automation: Another key part of vocal mixing, back in the day it was common practice for a sound technician to be something of a performer themselves, rehearsing and then live adjusting a gain lever response to keep levels uniform manually.
This has become something of a lost art in the age of DAWs- producers nowadays can simply view the waveform of their recording and apply an automation level to any aspect of their mix. Both can be used to reduce unwanted frequency occurrences in your recording.