EsscapeDe-esser · harsh esses softened, the rest of the voice left as it was
The only de-esser plugin you'll ever need to soften every harsh S.
Harsh S, SH and T sounds are turned down the moment they fire, and the rest of the voice is left as it was. The threshold follows your track's level, so one setting does the same job on a quiet take and a loud one, with zero latency.
- Use it on 5 computers per copy
- Lifetime updates
- 90-day money-back guarantee
- Download link in minutes
- VST3, AU and AAX, Windows and macOS
- Try it free
Formats and DAWs
Windows 10 and 11 (64-bit): VST3 and AAX
macOS 11 or later: VST3, AU and AAX
Works in any 64-bit DAW that loads VST3, AU or AAX plug-ins.
Licence
Use it on up to five of your own computers per copy you buy. Yours for good, no subscription. Unlock it with your email and a six-digit code, no dongle and no password. It checks in now and then and works offline in between.
Delivery
Download links by email within about five minutes, and always in your account under Downloads.
Hear the difference
The same vocal, without and with Esscape.
Sharp esses softened on a dry pop vocal
A dry sung pop refrain - Init (AMOUNT 50 %, VOICE HIGH, SPEED MEDIUM)
Hard S and T sounds tamed on a rap verse
A rap verse with quick, hard esses - Rap / Hot Rap (AMOUNT 70 %, VOICE HIGH)
Spiky esses sit back into a bright vocal stack
A stacked, bright sung refrain - Singing / Heavy Esses (AMOUNT 85 %, VOICE HIGH)
Every ess softened on a bright drum and bass vocal hook
A bright sung hook full of esses - Singing / Bright Female (AMOUNT 55 %, VOICE HIGH)
Esses lose their edge on a soft, close ballad vocal
A soft, close sung ballad - Singing / Gentle Pop (AMOUNT 40 %, VOICE HIGH), a light touch
Clean echo repeats: the esses go before the echo
A sung verse into a dotted-eighth echo - Mix / Before the Echo (AMOUNT 60 %, VOICE HIGH); same echo on both
Every pair is a sung or rapped vocal, loudness-matched; on the last one the same echo follows on both sides.
Esscape in action
Esscape in under a minute on a synthesised sung phrase: AMOUNT, the meter, VOICE, SPEED and LISTEN.
Why Esscape
Only the esses
Esscape turns down only the top of the voice, and only while an ess sounds. Between esses the signal passes through exactly, and below 800 Hz the default moves the voice by 0.04 dB, so nothing turns dull.
The same job at any level
The threshold follows your track's loudness: the default takes about 3 dB off the loudest esses at -10, -18 or -26 LUFS alike, where a fixed threshold would cut 4.4, 2.4 or 0 dB.
Set it by ear
VOICE says in plain words where the esses sit, LISTEN plays only what is being removed, and the REDUCTION meter reads on the 2-6 dB scale mixing guides use.
Soften the esses. Keep the voice.
AMOUNT does the work: how far the esses come down, with what that means printed under the knob, from LIGHT to HEAVY.
- VST3 · AU · AAX
- Windows and macOS
- Zero latency
- 14 presets
The de-esser plugin
- Only the esses. Between esses the signal passes through exactly; below 800 Hz the default moves the voice 0.04 dB.
- Any level. About 3 dB off the loudest esses at -10, -18 or -26 LUFS alike.
- Set it by ear. VOICE in plain words, LISTEN to hear only what is removed, and a REDUCTION meter on the scale mixing guides use.
- Zero latency. Record through it, and put it before any reverb or echo so the repeats stay clean.
- AMOUNT, VOICE (LOW, HIGH, AIR) and SPEED (FAST, MEDIUM, SLOW).
- LISTEN, the REDUCTION meter, MIX for parallel de-essing, Input and Output gain, IN and OUT meters, and ENGAGE.
- 14 factory presets for singing, rap, speech and mixing, and your own preset files; A/B, undo and redo, MIDI Learn on every control; a window from 75 to 200 %; the full manual.
AMOUNT · VOICE
The same job, at any level.
The threshold follows your track's loudness, so the default takes about 3 dB off the loudest esses on a quiet take and a loud one alike.
What Esscape does
Every control has one job: softer esses, with the rest of the voice left as it was.
AMOUNT
How far the esses come down
Moves the threshold and the deepest cut together. About 1.4 dB off the loudest esses at 25 %, 3 dB at 50 % (the default), 4.7 dB at 75 % and 5.5 dB at 100 %; at 0 % the voice is untouched. The line under the knob reads OFF, LIGHT, SMOOTH, FIRM or HEAVY.
Follows the level
The same job on every take
The threshold follows the track's loudness, so the default takes about 3 dB off the loudest esses at -10, -18 or -26 LUFS. A fixed threshold would cut 4.4, 2.4 or 0 dB: too much on the loud take, nothing on the quiet one. Turning IN does not change the de-essing.
VOICE
Where the esses sit, in plain words
LOW, 3-6 kHz, for deeper and male voices; HIGH, 5-8 kHz, for female and bright voices and most others; AIR, 8-12 kHz, for a brittle, hissy top. A change glides over 50 ms, so it never clicks, even in the middle of an ess.
SPEED
How the cut lets go
FAST, MEDIUM or SLOW: back under 1 dB about 10, 40 or 97 ms after an ess. FAST for quick, dense rap; SLOW so long notes and speech never pump. Each is calibrated to change how the cut lets go, not how much.
LISTEN
Hear exactly what goes
Plays only what is being removed. With the right VOICE and AMOUNT you hear a string of S sounds and none of the vowels. It switches with a 20 ms crossfade, is never saved in a preset and is never an undo step.
REDUCTION meter
See every ess come down
The cut right now, as heard in the middle of the VOICE band, on a 12 dB scale that reads like the 2-6 dB mixing guides give: about 4 dB on the loudest esses at the default. The number under it holds the deepest cut of the last 0.6 s.
Only while an ess sounds
The rest of the voice left as it was
Only the top of the voice is turned down, and only during an ess; between esses the signal passes through exactly. Below 800 Hz the default moves the voice 0.04 dB (the median of 17 test vocals).
Zero latency
Record through it
Nothing looks ahead, so it adds no latency at any setting. Monitor through it while you record, and put it before any reverb or echo so the repeats never spit every S back at you.
MIX
Soften the esses, keep some edge
A straight blend of the de-essed and the dry signal, exactly in time, so nothing phases. Around 50 % the esses soften but keep some bite, as Mix / Parallel Soften does.
IN and OUT
Fit it to the track
IN and OUT, -24 to +24 dB with meters, set the level either side, and a safety stage keeps the output under -1 dBFS. A de-esser only takes the esses away, so no preset touches OUT.
14 sourced presets
Start from a documented setting
Singing (Gentle Pop, Bright Female, Low Male, Air Tamer, Heavy Esses), Rap (Hot Rap, Low Rap), three Spoken presets on the slow release, and Mix (Before the Echo, Light Touch, Parallel Soften), each within 1 LU of the dry on average.
A window that breathes
See each ess lift the feather
A white feather on the sill lifts as the esses are cut, reaching the top of its rise at 7 dB, and drifts back down as the cut lets go. The moon's halo glows with the output; at AMOUNT 0 % or with ENGAGE off the feather lies still.
Inside the plugin
The Esscape window: AMOUNT, the VOICE and SPEED keys, the REDUCTION meter and a feather that lifts with every ess.
Every feature
Everything in Esscape, down to the smallest switch.
37 features
The de-essing
-
AMOUNT
How far the esses come down, from OFF to HEAVY, printed under the knob
-
Follows the track's level
About 3 dB off the loudest esses at -10, -18 or -26 LUFS alike
-
VOICE
LOW 3-6 kHz, HIGH 5-8 kHz or AIR 8-12 kHz: where the esses sit, in plain words
-
SPEED
FAST, MEDIUM or SLOW: lets go before the next syllable, or never pumps
-
Only while an ess sounds
Between esses the voice passes through exactly
Hear it and see it
-
LISTEN
Hear only what is being removed, to set AMOUNT and VOICE by ear
-
REDUCTION meter
The cut right now on a 12 dB scale, and the deepest of the last 0.6 s
-
A window that moves
The feather lifts with every ess cut; the moon glows with the output
Timing and mix
-
Zero latency
Record and monitor through it at any setting
-
MIX
Parallel de-essing, exactly in time, so nothing phases
-
Input and Output gain
-24 to +24 dB, with a safety stage under -1 dBFS
-
Level-matched presets
Every preset within 1 LU of the dry on average
Workflow and control
-
A/B compare and A>B
Two complete settings, switch in one click
-
Undo and redo
Step back and forward, or press Ctrl+Z
-
MIDI Learn
Take complete control of your workflow: right-click any control, move a knob
-
Host automation
Automate every control from your DAW
-
ENGAGE
Click-free bypass, to hear the before and after
-
IN and OUT meters
See the level going in and coming out at a glance
-
Tooltips
Every control explains itself as you point at it
-
Resizable window
75 to 200 %, and the size you pick is remembered
Presets
-
14 factory presets
Each one built from a documented setting
-
Previous and next arrows
Audition presets one click at a time
-
Save your own
Keep your settings as .deesserpreset preset files
-
Open presets folder
Back up or share your presets in one click
-
Reset to default
Back to a clean start in one step
Performance
-
Mono or stereo
Each channel de-essed on its own; mono works the same
-
Light on CPU
About 0.7 % of one core for the sound at 44.1 or 48 kHz, stereo
-
Any session rate
Bands and release times tested at 44.1, 48 and 96 kHz
-
Any buffer size
Identical output at 128, 512 and 777-sample blocks
Licence and compatibility
-
VST3 and AAX, Windows 7 SP1 or later
Loads in any 64-bit Windows DAW with VST3 or AAX
-
VST3, AU and AAX, macOS 10.13 or later
Loads in any Mac DAW with VST3, AU or AAX, Intel or ARM
-
5 computers per copy
Studio, laptop and more on one licence
-
Email and code unlock
No dongle, no password, no extra app
-
Works offline
No internet needed between check-ins
-
Check for updates
From the plugin's menu; nothing installs without asking
-
Lifetime updates
Every future update included
-
Platinum priority support
Questions answered by the people who build it
How Esscape compares
Esses softened by the same amount on any take, set in plain words, with the checks most leave out.
| Esscape | Typical de-esser plugins | |
|---|---|---|
| Threshold | Follows the track's level: one setting for quiet and loud takes | Often fixed: misses quiet takes, over-cuts loud ones |
| Finding the esses | VOICE in plain words: LOW, HIGH or AIR | Often a frequency knob to sweep and guess |
| Checking the cut | LISTEN, and a meter on the 2-6 dB scale mixing guides use | Often a number with no way to hear it |
| Rest of the voice | Passed through exactly between esses | Often an EQ cut that dulls it all the time |
| Latency | Zero at every setting: record through it | Sometimes look-ahead delay |
| Try-and-compare | A/B, A>B, undo and redo | Often missing |
| Hardware control | MIDI Learn on every control | Often host automation only |
| Licence | 5 computers per copy, no dongle, works offline | Often fewer activations, or a dongle |
| Updates | Lifetime updates included | Often paid upgrades |
"Typical" describes what we commonly see in this category; it does not describe any particular product. Esscape facts are from the plugin and its manual, October 2026.
What you get
Vox Samples
90-day money back · Thank you
Specs and system requirements
- Formats
- VST3 and AAX on Windows; VST3, AU and AAX on macOS
- Operating system
- Windows 7 SP1 or later, 64-bit; macOS 10.13 or later, universal (Intel and ARM)
- Host
- Any 64-bit DAW that loads VST3, AU or AAX plug-ins. No standalone app.
- Channels
- Mono or stereo; each channel de-essed on its own (no stereo link)
- Sample rates
- Whatever your session runs at; tested at 44.1, 48 and 96 kHz
- Latency
- 0 samples at every setting (no look-ahead)
- Bands
- VOICE: LOW 3-6 kHz, HIGH 5-8 kHz, AIR 8-12 kHz
- Release
- SPEED: FAST, MEDIUM, SLOW, time constants of 5, 20 and 50 ms; back under 1 dB in about 10, 40 and 97 ms
- Cut
- About 1.4, 3.0, 4.7 and 5.5 dB on the loudest esses at AMOUNT 25, 50, 75 and 100 %
- Meter
- REDUCTION, 12 dB full scale, plus the deepest cut of the last 0.6 s
- CPU
- About 0.7 % of one core for the sound at 44.1 or 48 kHz, stereo; the open window adds about 1.1 to 1.2 %
- Presets
- 14 factory presets; your own as .deesserpreset files
- Manual
- https://voxsamples.com/esscape-de-esser-manual
- Window
- 960 x 600 at 100 %, resizable from 75 to 200 %
- MIDI
- MIDI Learn from any controller routed to the track
- Licence
- 5 computers per copy you buy. Unlock with your email and a code we email you. Checks in now and then and works offline in between.
- Internet
- Needed once to unlock, then for the occasional check-in
90Day money-back guarantee
Our guarantee
-
90-day money-back guarantee
Not for you? Email contact@voxsamples.com within 90 days for a full refund.
-
5 computers per copy
Use it on up to five of your own computers per copy you buy, and move it yourself from the plugin's Licence menu.
-
Lifetime updates
Every future update and upgrade of Esscape, included.
-
Platinum priority support
A real person at contact@voxsamples.com, first in line.
Try Esscape free
The whole plugin, with the same sound as the full version and no time limit. It goes silent for a moment every few seconds; the full version never does.
Windows and macOS installers in one download. Like what you hear? Get the full version on this page, with a 90-day money-back guarantee.
Questions
What does a de-esser do?
It turns down the harsh S, SH and T sounds in a voice when they happen, without dulling the rest of the voice the way an EQ cut would.
Where do I put it?
On the vocal track, after any EQ or compression and before any reverb or echo, so the effects never get the esses.
How much should it cut?
The default takes about 3 dB off the loudest esses, inside the 2-6 dB range mixing guides give. Raise AMOUNT until the esses stop jumping out; back it off if the voice sounds dull or lispy.
Which VOICE do I pick?
HIGH for most voices, LOW for deeper and male voices, AIR for a brittle, hissy top. With LISTEN on, the right one lets you hear the esses and none of the vowels.
Will it make my vocal quieter?
Slightly: it takes the esses' energy away and does not make it up. About 0.4 LU on average at the default; very bright or hissy takes lose up to about 2 LU.
Does it add latency?
No. Zero at every setting, so you can record and monitor through it.
Does it work on speech?
Yes. The three Spoken presets use the SLOW release, so long reads never pump.
Can I use it on a stack of vocals?
Yes: one Esscape on the group bus treats the whole stack together. Singing / Heavy Esses is for a stack whose esses build up.
Does it work in my DAW?
Any 64-bit DAW that loads VST3, AU or AAX plug-ins, on Windows 7 SP1 or later or macOS 10.13 or later. There is no standalone app.
How many computers can I use it on?
Up to five of your own for each copy you buy. To move it, deactivate it in the plugin on the old computer and unlock it on the new one.
Does it need the internet?
Once, to unlock it with the code we email you. After that it checks in now and then and works offline in between.
Is there a manual?
Yes. The full manual explains every control, every preset and the menus, with tips. It opens from the help key (?) in the plugin's top bar, and lives at https://voxsamples.com/esscape-de-esser-manual.
What if it's not for me?
Request a refund from your account within 90 days.
Founder's Story
"When COVID hit, I finally had time to dive into music production—only to discover I was missing the right tools. The plugins I tried were clunky, overloaded with unnecessary features, and killed my vibe with endless tweaking. I knew there had to be a better way.
So, with my engineering background, I started making free and simple audio plugins for my producer friends. They loved them! That’s when I set out on a mission to create plugins that save you time, streamline your workflow, and keep your creativity flowing."
- Ryan Svensson (Founder & CEO)
Harsh esses softened. The voice left as it was.
Esscape turns down only the esses, only while they sound, by the same amount on a quiet take or a loud one, with zero latency and LISTEN to hear exactly what goes.
$42.55 AUDPrecio habitual $141.86 AUD70% off
- 90-day money back
- Five computers
- Lifetime updates
Pairs with
Esscape De-Esser
$42.55 AUDPrecio habitual $141.86 AUD70% off