How To Use De-Essers For Vocal Production/Mixing
De-essers are indispensable tools in the realm of vocal production. They help in achieving a balanced and professional mix by controlling harsh sibilance. Whether you're a seasoned producer or just starting, incorporating a de-esser into your vocal processing chain is essential for producing clear, polished vocal tracks.
What is a de-esser? A de-esser is a dynamic audio processor that selectively reduces sibilance ( ‘s' or ‘z' sounds, typically in the 4-8 kHz frequency range) in vocal tracks. It acts as a frequency-focused compressor, engaging only as the sibilances are accentuated to reduce harshness while maintaining clarity.
If you're wondering what a de-esser is (in more detail than the short answer above), how they work, and how we can use them to our fullest advantage in the mix, you've come to the right article!
If you'd prefer the video format for learning about de-essers, please check out my video below:
What Is A De-Esser?
A de-esser is an audio processing tool primarily used in vocal production. Its main function is to reduce or eliminate sibilance, which are harsh “s” or “sh” sounds that can be overly prominent in vocal recordings.
It works similarly to a compressor but specifically targets the high-frequency range where sibilance typically occurs. By selectively compressing these frequencies, a de-esser helps to smooth out the vocal track, making it clearer and more pleasant to listen to without affecting the overall tonal balance of the voice.
How Does A De-Esser Work?
So we know the basics of why we'd want to use a de-esser: to dynamically control the sibilance of, most typically, a vocal.
From that short description, we know that it's a dynamics processor. In case you're unaware, the most common type of dynamics processor is the good old compressor. Therefore, if we understand compression, we can understand de-essing.
Please forgive the upcoming jargon if you're new to it all. I've included links to more articles so that you can learn what you need to know to gather a complete picture of how compression and de-essing work.
To keep things as concise as possible, a compressor works to reduce the dynamic range of a signal (the difference between the loudest and quietest parts of the audio). A compressor will have two main signals:
- Program signal: this is the audio being compressed. It goes into the compressor via the input, is processed by the gain reduction circuit/algorithm (it gets compressed), and may have additional makeup gain applied before it's outputted.
- Sidechain signal: this is the signal that controls the gain reduction circuit/algorithm (the amount of compression). While the sidechain is often taken from the input (either immediately before or after the gain reduction circuit/algorithm), it can also be an external signal — this is where the term/technique of “sidechain compression” comes from.
The way a compressor will respond to the sidechain and, therefore, dynamically reduce the gain of the program signal is dependent on the following parameters:
- Threshold: a level set by the user. When the audio signal exceeds this level, the compressor becomes active. Any part of the signal below the threshold remains unaffected.
- Ratio: the amount of gain reduction applied to the signal once it crosses the threshold. For example, a 4:1 ratio means that for every 4 dB that the input signal exceeds the threshold, the output signal will only increase by 1 dB.
- Attack time: how quickly the compressor starts to work after the signal exceeds the threshold. A fast attack time means the compressor reacts quickly, while a slow attack time allows some of the signal's initial transient to pass through uncompressed.
- Release time: how quickly the compressor stops acting after the signal falls below the threshold. A fast release time quickly stops compression, while a slow release time maintains compression for longer.
Other parameters are involved, but I'm trying to keep this “simple”.
So, with that information, let's consider how a de-esser works as a dynamic processor:
A de-esser provides frequency-selective compression, aimed, as you could guess, at reducing the harshness of sibilant sounds (like “s” or “sh” sounds).
In order to work effectively, de-esser uses sidechain processing, where the compressor is triggered not by the overall level of the audio signal but rather by a filtered version of the input. This sidechain signal is generally either high-passed or band-passed.
- High-pass filters (HPFs) pass frequencies above a set cutoff/corner frequency and remove content below the cutoff/corner frequency. Therefore, when setting an HPF appropriately in the de-esser sidechain, we're causing the most sibilant frequencies and all the frequencies above to trigger the gain reduction.
- Band-pass filters (BPFs) only pass frequencies in a specified band, typically with a set centre frequency, and remove frequencies outside this band. Therefore, when setting a BPF appropriately in the de-esser sidechain, we're causing only the most sibilant frequencies to trigger the gain reduction.
I should mention that some de-essers do indeed allow for an external sidechain signal, but we'll assume, for simplicity's sake, that we're taking the sidechain signal from the input.
Beyond the triggering of the de-esser's gain reduction, we also typically have control over what frequencies are actually affected by the de-esser. Generally, the two options are split-band or wide-band:
- Split-band mode target only the sibilance range and are most often linked to whatever sidechain filtering we used. In other words, if we high-passed the sidechain, the de-esser would act only upon those high-passed frequencies, and if we band-passed the sidechain, the de-esser would act only upon the band-passed frequencies.
- Wide-band mode targets the entire signal across the frequency spectrum. Whenever the de-esser applies any amount of gain reduction, it does so equally to all the frequencies.
Like in compressors, users can generally set a threshold level. When the sibilant frequencies exceed this threshold, the de-esser reduces their volume. The sensitivity or intensity of this effect can usually be adjusted.
There generally isn't a ratio control in de-essers. However, they sometimes have a range or maximum reduction control that we can set so that the de-esser never applies more than the set maximum gain reduction.
Generally, a de-esser is designed with quicker attack and release times compared to a standard compressor. This allows it to swiftly diminish the brief, sharp transients of sibilance in vocals, effectively minimizing them without noticeably affecting the rest of the audio signal, either preceding or following the sibilant sounds.
Some advanced de-essers have a lookahead feature, allowing them to react almost instantaneously to sibilant sounds, making the process more effective and transparent.
Those are the “basics” of how a de-esser works. Again, please feel free to follow any of the links above for more information on anything that you may need clarification on (or ask a question in the comment section below).
Why Use A De-Esser?
De-essers are used to dynamically control these harsh sibilance frequencies without affecting the overall character of the vocal.
Unlike static EQ adjustments, de-essers work dynamically, only engaging when sibilant sounds are detected. This selective processing ensures that the natural quality of the vocal is maintained while taming the harshness.
A de-esser is a dynamic audio processing tool specifically designed to reduce or eliminate harsh sibilant sounds commonly found in vocal recordings.
These sibilant frequencies typically range from 4,000 to 10,000 Hz and can be overly prominent in a mix, leading to a distracting and fatiguing listening experience.
It's actually the case that human hearing is the most sensitive around 4 kHz and that a large part of speech intelligibility is dependent on these upper-midrange frequencies.
So they're important, yet too much energy in this range will lead to harshness and rapid ear fatigue.
I'm discussing vocals almost exclusively here because de-essers are primarily used to treat vocals. However, I felt it necessary to add that de-essers can also be effective on other types of audio signals. Here are some common applications:
- Cymbals and hi-hats in drum tracks: de-essers can tame the harshness or overly bright sounds of cymbals and hi-hats, especiallyin cases where they dominate the mix or sound too piercing.
- String instruments: for stringed instruments like violins or guitars, a de-esser can help control sharp, high-frequency sounds that might occur due to the picking or bowing technique or while sliding the fingers along the strings.
- Wind instruments: instruments like saxophones or trumpets can sometimes produce high-frequency hissing or piercing sounds. A de-esser can smooth these out for a more balanced sound.
- Synthesizers and electronic sounds: certain synthesizer patches might have harsh high frequencies. A de-esser can be used to soften these without affecting the lower frequencies of the patch.
Personally, I'll generally reach for a dynamic EQ to make such adjustments to the cymbals, strings, woodwinds, synthesizers, etcetera that require some additional dynamic attenuation in specific frequency bands, but de-essing can work, too.
How To Use A De-Esser
Alight, we know what de-essers are, why we should use them, and how they work on a more technical level. Now, let's go through the steps of actually putting them to use.
Before we get started, remember these important points when it comes to all processing, including de-essing:
- Use your ears: always trust your ears more than the visual feedback from the plugin.
- Less is sometimes more: subtlety is often key with de-essing. Over-de-essing can make the vocals sound unnatural and “lispy”.
- Monitor on different systems: check the de-essed vocal on different playback systems to ensure it sounds good universally. This is important for the entirety of the mix in terms of translatability.
Also, note that different de-essers will have different numbers of controls and their controls may be labelled differently. I'll do my best to add whatever terms may apply to each parameter in question.
Alright, here's an in-depth, step-by-step guide to using a de-esser:
1. Identify The Problem Areas
Before applying a de-esser, listen carefully to the vocal track to identify sibilant frequencies. You might even want to sweep a parametric EQ boost to help identify the sibilant issues. These are usually high-frequency sounds, typically between 4 kHz and 8 kHz, but can vary depending on the voice.
2. Insert The De-esser
In general, I advise inserting the de-esser after any EQ or compression in your vocal chain. This placement ensures that the de-esser acts on the processed signal, which might have enhanced sibilance due to EQ boosts or compression.
3. Set The Frequency
Use the de-esser's frequency control to pinpoint the exact frequency range where the sibilance is most prominent. Some de-essers allow you to solo the targeted frequency band, making it easier to identify.
4. Adjust The Threshold
Set the threshold so that the de-esser engages only when sibilant sounds occur. If set too low, it will affect the overall clarity of the vocal; if set too high, it won't catch the sibilance.
The key is to adjust the threshold and frequency settings to ensure that only the harsh sibilant sounds are being reduced without affecting the overall clarity and brightness of the vocal.
5. Choose The De-essing Method
Some de-essers offer a choice between ‘wide-band' and ‘split-band' processing. Wide-band reduces the level of the entire signal when sibilance is detected, while split-band only reduces the level of the specific frequency range. Split-band is often more transparent, but can sound much less natural if it ever does become overly apparent in the mix.
6. Adjust The Range
This parameter controls how much the sibilant frequencies are reduced. Adjust it to achieve a balance where sibilance is controlled without making the vocals sound lifeless or lispy.
7. Listen In Context
Always listen to the de-essed vocal in the context of the full mix. Sometimes what sounds overly sibilant in solo can be perfectly fine in the mix.
8. Fine-Tuning
You may need to go back and forth, adjusting the threshold, frequency, and range to get the perfect balance. Pay attention to how the de-esser affects the overall tonality of the vocal.
9. Use Automation If Necessary
In some cases, you might need to automate the de-esser's parameters for different sections of the song if the sibilance varies throughout the performance.
10. A/B Comparison
A/B test the de-esser regularly by bypassing it to compare the before and after. This check ensures that you're actually improving the sound and not over-processing it.
11. Additional EQ
Sometimes you might need to apply a slight EQ boost after de-essing to restore some of the airiness or presence that the de-esser might have attenuated. I'll actually often reach for the classic Plugin Alliance Mäag EQ4 to add some “Air Band” if the de-esser sucks too much life out of the vocal's top end.
By following these steps and tips, you can effectively use a de-esser to enhance vocal clarity and quality in your productions, ensuring a professional and polished final product.
My Favourite De-Essers
In the early days of my production journey, I never de-essed the vocals in my mixes (or tuned them, for that matter). I was completely under the impression that not processing my vocals would give them that raw character I liked as if it was something to brag about. Looking back, I was completely wrong, and my vocal production suffered because of it. Don't be like me in the early days!
Since then, I've used quite a few different de-essers over the years. The ones I have the most experience with are also my favourites (funny how I choose to use my preferred tools).
They are the FabFilter Pro-DS, the Waves DeEsser, and the trusty Logic Pro X's stock DeEsser 2. Each of these tools offers unique features and controls to handle sibilance effectively.
Let's look at them in a bit more detail:
FabFilter Pro-DS
The FabFilter Pro-DS is known for its user-friendly interface and precise control. To me, it stands out for its precision and flexibility.
It features highly intelligent and transparent algorithms that adeptly target sibilance, ensuring that the de-essing process is both effective and unobtrusive. This plugin offers a unique ‘Single Vocal' detection mode, which is specifically optimized for vocal de-essing, making it incredibly efficient at honing in on sibilant frequencies without impacting the integrity of the overall vocal performance.
Additionally, its user-friendly interface and visual feedback allow for precise control, enabling users to fine-tune settings to suit specific needs.
It offers a ‘Range' knob to set the maximum gain reduction and a ‘Threshold' control to determine when the de-esser engages.
The frequency range for de-essing can be adjusted.
It features a ‘Split Band' mode to target only the sibilance range and a ‘Wide Band' mode that affects the entire signal.
The Pro-DS's versatility extends beyond vocals, as it can be effectively used on a wide range of audio sources, making it a top choice for both music producers and audio engineers seeking a high-quality, versatile de-essing tool.
Waves DeEsser
The Waves DeEsser is another great choice, and I really like it for its effectiveness and user-friendly design.
Indeed, one of the key strengths of the Waves DeEsser is its simplicity, making it my personal go-to. It offers intuitive controls that make it easy for both beginners and professionals to quickly achieve the desired de-essing effect.
The plugin operates with a clarity that preserves the natural tone and character of the voice, ensuring that the de-essing process doesn't detract from the overall vocal quality. Additionally, its versatility allows it to be used effectively on a variety of audio sources beyond vocals, such as cymbals or other high-frequency instruments.
The DeEsser allows for ‘Audio Split' or ‘Wide' mode, offering flexibility in how the gain reduction is applied.
It includes a ‘Frequency' control to target specific sibilance frequencies.
The sidechain monitoring feature helps in precisely identifying the sibilant frequencies.
Logic Pro X DeEsser 2
This stock plugin has a straightforward and intuitive interface, making it accessible to users of all skill levels. It offers the key controls, including threshold, frequency selection, a ‘Max Reduction' control for limiting the amount of gain reduction, and a ‘Filter Solo' option for isolating the sibilance frequencies.
And the best part is that it's free with Logic Pro X! Its seamless integration with Logic Pro means it operates efficiently within the DAW, utilizing minimal CPU resources. This makes it an excellent choice for Logic Pro users looking for a reliable, easy-to-use de-essing tool that delivers professional results.
As a Logic Pro X user, I've had plenty of practice with the DeEsser 2, and while it may not necessarily be my go-to choice, it's a great de-esser in its own right.
Call To Action!
If you're unfamiliar with de-essing, take some time to experiment and get comfortable with whatever stock de-esser comes with your DAW, or choose a third-party option. Push things too far and then dial them back to gain a better understanding of what de-essers do and how they make your vocals sound.
Have any thoughts, questions or concerns? I invite you to add them to the comment section at the bottom of the page! I'd love to hear your insights and inquiries and will do my best to add to the conversation. Thanks!
Related Questions
What are other tips for producing better vocals? Here are 11 more tips for better vocal production:
- Recording dry
- Comping and multing
- Time-aligning background vocals
- Doubling harmonies
- Doubling lead vocals
- Vocal tuning
- Compression
- Intricate volume/fader automation
- Parallel processing
- Adding space with delay and reverb
- Delay and reverb throws
Related Article On Producing Better Vocals
To learn more about vocal production, check out my article How To Produce Better Vocals (12 Pro Tips)—de-essing is on the list!
What are the best microphones for recording vocals? In my opinion, the following microphones are the top of the line for recording vocals. They're expensive, but they're awesome:
- AKG C 12
- Telefunken Ela M 251 E
- Neumann U 67
- Neumann U 47
- Sony C-800G
- Manley Reference Gold
- Neumann U 87
- Neumann TLM 103
- AKG C 414
- Coles 4038
- Shure SM7B
