How to Sound Good (Part 5) - Tone Shaping and De-Essing

This is Part 5 of my “How to Sound Good” series.

If you have been following along and building your audio chain step-by-step, you are in a pretty great spot right now. You bought a decent microphone ( Part 1 ), you gated the silence ( Part 2 ), you scrubbed the background noise ( Part 3 ), and you used a compressor and limiter to make your volume consistent and safe ( Part 4 ).

Your audio is now clean, quiet, and reliably loud. You have solved the technical problems.

But you might still sound a little… weird.

Maybe your voice sounds “boxy,” like you are broadcasting from inside a cardboard tube. Maybe the proximity effect of your microphone is making you sound too bass-heavy and muddy. Or maybe—and this is the most common side effect of compression—your “S” and “T” sounds are suddenly piercing enough to shatter glass.

It is time to shape the actual tone of your voice.

Before we begin: Accept that you’ll always sound a little weird

Have you ever noticed how when you listen to a recording of yourself, it never sounds like “you”? It doesn’t quite match the voice you hear when you’re speaking. It might be close, but it’s just not the same!

That’s not you. That’s anatomy. What you hear in your head is a combination of sound waves propagating through the air, combined with sound waves propagating through your body. Your vocal cords are rattling your blood, muscle, and bones, and you’re always hearing that quite loudly, right alongside the sounds traveling through the air. The result is that no matter what you do, your voice will always sound different to you than it does to others.

Sure, you could tweak and modify your sound until what people hear in meetings really does match what you hear in your head. But then, to them, you’ll sound completely different in-person, because what they are hearing in meetings is fundamentally a little weird.

The moral here is to just not sweat it. The goal isn’t for others to hear what you hear when you talk. The goal is for you to simply sound good. You want to sound like you do when others hear you, but with a bit of polish. So don’t judge recordings of yourself against what you think you should sound like. Instead, treat it almost like you’re working on someone else’s voice, and you just want them to sound their best.

A good mindset to have (for this, along with pretty much everything else covered in the previous posts in this series) is that you aren’t trying to actively change or improve your sound. Instead, you want to find the problems and fix them, and just trust that your voice is naturally good and stands on its own. You are just helping it a little along the way.

The Sculptor’s Chisel: Equalization (EQ)

An Equalizer (EQ) is basically a volume knob, but instead of turning down the entire audio signal, it lets you turn down specific frequencies.

If you remember the old stereo systems with sliders for “Bass,” “Mid,” and “Treble,” that is a rudimentary EQ (referred to as a 3-band EQ). Modern plugins give us a visual graph (a Spectrum Analyzer) where we can literally see the low frequencies on the left, the high frequencies on the right, and drag them up or down with surgical precision.

For remote work, we aren’t trying to make you sound like a movie trailer voiceover (unless that’s your natural voice). Our goal is intelligibility. We want to carve away the frequencies that make you hard to understand.

The 4 Essential EQ Moves for Voice

When I set up a new vocal chain in Bitwig, I almost always make these exact moves before I even listen to the mic.

1. The Mud Cut (High-Pass Filter) Your voice does not produce any useful information below roughly 80Hz. However, your microphone will happily pick up the low-frequency rumble of a passing truck, the hum of an air conditioner, or the thud of your hand hitting the desk.

• The Move: Activate a High-Pass Filter (HPF, also known as a low-cut filter) and drag it to somewhere between 80Hz and 100Hz. This immediately cleans up the low-end “mud” and gives your compressor less garbage to react to.

2. The Boxy Scoop Small, untreated rooms (like a home office or a motorhome) create nasty, short reflections that build up in the low-mid frequencies. This buildup makes you sound like you have a head cold.

• The Move: Create a bell-shaped curve around 300Hz to 500Hz and pull it down by -2dB to -4dB. You will instantly sound less congested. I also like to lower the “Q” parameter on this one, to widen it out and smoothly cover a pretty wide range.

3. The High Cut (Optional) Most people who aren’t children can’t hear much above 15 kHz, and there’s just not much useful audio information there anyway, so most meeting software doesn’t bother sending those higher frequencies over the wire (this saves a considerable amount of bandwidth and helps with the distortion from terrible microphones and crappy analog-to-digital converters). Those really high frequency sounds aren’t doing you any favors with spoken words, and if your meeting software isn’t filtering them out anyway, they tend to be where a lot of background noise lives.

You can use a pretty steep filter for this, which helps keep it from messing with your lower-frequency stuff too much.

• The Move: Add a low-pass (also known as a high-cut) filter around 15-16 kHz to remove those frequencies entirely for a cleaner sound.

4. The Air Boost (Optional) If your microphone sounds a little dark or muffled, you can add some artificial clarity.

• The Move: Create a “High Shelf” at around 8kHz and boost it by +1dB or +2dB. This adds presence and “shimmer,” making you sound more like a professional broadcaster. You can do this in a separate instance of your EQ plugin, so that you can use it as a final “polish” after other processing steps.
• The Warning: Be very careful here. Boosting high frequencies will make your “S” sounds much harsher, which leads us directly to the next tool.

My EQ Recommendations

• The Free Option: TDR Nova . It has a built-in spectrum analyzer so you can see your voice, and it functions as a dynamic EQ (which we will talk about in a second). It is shockingly good for $0.
• The Premium Options: If you catch it on sale, Three-Body Technology Kirchhoff-EQ is a powerhouse. If you hate money and want the absolute gold standard of workflow, FabFilter Pro-Q 4 is the industry standard (if you’ve listened to just about any music made in the last decade, they almost certainly used Pro-Q).

Taming the Snakes: The De-Esser

Here is a frustrating irony of audio engineering: The things you do to fix one problem usually create another.

When we used a compressor in Part 4 to make your quiet words louder, we also accidentally made your mouth noises louder. Condenser microphones are already highly sensitive to high frequencies. When you compress that signal, the sharp bursts of air from consonants like S, T, and Ch (called sibilance) get pushed to the front of the mix.

To the people listening to you on headphones, those “S” sounds can be physically painful.

To fix it, we use a De-Esser.

How it Works

A de-esser is essentially a very fast, very stupid compressor. It only listens to the high frequencies (usually between 5kHz and 9kHz). When it hears a massive spike in that specific zone, it rapidly turns the volume down for a fraction of a second, and then lets go.

It squashes the “S” before it stabs your coworker in the ear.

The De-Essing Strategy

Setting a de-esser is a delicate balancing act.

• Too little: You still sound harsh and piercing.
• Too much: You turn the “S” into a “Th” and you suddenly sound like you have a severe lisp.

The Move: Lower the threshold until you notice your “S” sounds softening. If you start to lisp, dial it back slightly.

My De-Esser Recommendations

• The Free Gem: AnalogObsession LOADES . This is a phenomenal, free plugin. Turn on the “Soft” button, lower the threshold, and enjoy incredibly transparent smoothing.
• The “All-In-One” Strip: If you use something like SSL VocalStrip 2 , it has a highly capable de-esser built right into the interface.
• The Dynamic EQ Hack: Remember TDR Nova from the EQ section? It is a dynamic EQ, meaning you can tell a specific EQ band to pull down the volume only when that frequency gets too loud. You can easily use a dynamic EQ band focused at 7kHz to act as a highly surgical de-esser.

The Final Signal Chain

So, you have a pile of virtual gear. Where does it all go? In digital audio, the order of your plugins (the signal flow) drastically changes how they sound.

Here is how I structure my master vocal chain for remote work:

1. The Source: Microphone (Zoom H1essential)
2. Noise Reduction: (Supertone Clear or Klevgrand Brusfri). Clean the raw signal first.
3. The Gate: Silence the background when you aren’t speaking.
4. Subtractive EQ: (The High-Pass Filter and the Boxy Scoop, and maybe the High-Cut Filter). Remove the garbage frequencies so your compressor doesn’t overreact to them.
5. The Compressor: Smooth out your volume dynamics.
6. The De-Esser: Tame the harsh high frequencies that the compressor just made louder.
7. Additive EQ (Optional): Add that tiny high-end “Air” boost to bring the sparkle back after de-essing.
8. The Limiter: The brick-wall safety airbag to protect against accidental digital clipping.

The Verdict

If you build out this pipeline, you are no longer just “using a microphone.” You have built a broadcast-ready vocal chain.

It takes a little bit of time to route your audio through a host like Bitwig Studio and dial in these parameters. But once you save that template, you never have to think about it again. You just open your laptop, join the meeting, and let the software automatically sculpt, polish, and control your sound.

I’m thinking about doing a deeper dive into the actual software routing—how I pipe the audio from Bitwig into Zoom without causing a feedback loop that destroys the space-time continuum. Would a visual diagram of the macOS routing matrix be helpful, or is that getting a bit too far into the weeds?