User Manual
Clear Voice LIVE
AI · Voice · Processor
Version 0.9.5 · Beta
Made in Graz, Austria
© 2026 Fabio Schurischuster
1. Welcome
Clear Voice Live is a real-time AI voice processor designed for live sound,
broadcasting, conferences and streaming. It does four things at the same
time, on a single insert per vocal channel, at zero output latency:
- Denoising — removes background noise, fans, HVAC, audience bleed
- Feedback prevention — kills mic-monitor coupling before resonance builds up
- Dereverb — tightens room reflections and hall tails
- Voice presence — keeps the spoken/sung voice itself untouched
All processing is driven by an AI core that separates voice from background
frame by frame. There are no notch filters to draw, no rings to chase, no
static EQ.
2. Quick Start
- Install via the supplied
ClearVoiceLive-Setup.exe (Windows) or .pkg (macOS).
- Re-scan plug-ins in your DAW. Clear Voice Live appears under the Dynamics / FX category.
- Insert it on a single vocal channel (or on a vocal-group buss).
- Engine: SOFT. Strength: 0 %. Smooth: 1.0. Dereverb: 0 %.
- Increase Strength until the noise floor disappears without
losing voice character. If at high Strength the SOFT engine is still not
enough, switch to HARD and start at around 30 % Strength
from there.
That's the whole quick-start. The plug-in is intentionally
small — three faders, two safety buttons, two engines. Everything else in
this manual is detail you only need when you want to fine-tune.
💡 The three tips that matter most:
- Close the plug-in window during the show. The GUI
(meters, animations) costs extra CPU. With the window closed, every
instance runs in its minimal audio-only path — this is how the
"many instances at a few percent CPU" figures are achieved.
- SOFT is the low-CPU engine — use it everywhere.
HARD costs several times the CPU per instance and is not
simply "better" (see section 7). Run SOFT
on every vocal channel; reserve HARD for the one or two problem
channels that really need it.
- Always run a little Dereverb — even when you hear no
reverb. 1 – 3 on the fader measurably strengthens
feedback prevention on any stage.
3. System Requirements
| Operating system | Windows 10 / 11 (64-bit). macOS 11+ in progress. |
|---|
| CPU | x86-64 with SSE3. Apple Silicon native build planned. |
|---|
| Plug-in formats | VST3. AU and AAX in progress. |
|---|
| Host sample rates | 48, 96 and 192 kHz accepted. Other rates show a warning banner and bypass cleanly. |
|---|
| Internal processing | Always at 48 kHz, regardless of the host's sample rate (see section 4). |
|---|
| RAM | ~80 MB per instance. |
|---|
| Output latency | 0.0 ms (zero PDC reported to host). |
|---|
The plug-in is written so that the host can route mono or stereo channels
freely. Stereo input is processed in stereo, with stereo-linked gain
decisions to keep the image stable.
4. Audio Setup & Buffer Size
Clear Voice Live's internal latency is zero. The real latency you
hear in a live performance — the so-called roundtrip latency
— is determined by your audio interface, the host's buffer size, and the
driver running underneath. The plug-in itself adds nothing.
Roundtrip ≈ (driver IN) + (host buffer) + (plug-in chain) + (driver OUT)
Internal sample rate
The AI core inside Clear Voice Live runs at 48 kHz fixed
for all hosts. When the host runs at 96 or 192 kHz the plug-in does
internal sample-rate conversion before and after the analysis path. The
output stays at the host's rate. This means:
- There is no audio-quality benefit to running the host
at 96 or 192 kHz just for this plug-in — the AI work
happens at 48 either way.
- There IS a latency benefit, because the audio driver's
buffer is measured in samples, not milliseconds. A 64-sample
buffer at 96 kHz is 0.67 ms; the same 64 samples at 48 kHz
is 1.33 ms. Running the host at the highest rate the interface
supports comfortably halves driver latency.
Recommended driver setup
| Setting | Recommended | Why |
|---|
| Audio driver | ASIO (Windows), Core Audio (Mac) | Both are low-latency, professional drivers; avoid WASAPI / WDM in live use. |
| Host sample rate | The highest rate your interface runs without complaints — usually 96 kHz, sometimes 192 kHz. | Reduces driver latency for the same buffer-size setting (samples ÷ rate). |
| Buffer size | 32 – 64 samples | The smallest your interface and CPU can handle without crackles. See below. |
Choosing the buffer size
Clear Voice Live is a live plug-in. Anything above ~64 samples
introduces enough roundtrip latency that the artist can hear themselves
twice through the in-ear monitor — once acoustically through the head,
once electrically through the chain. That feels wrong and is unworkable
for live use.
- 32 samples: tightest possible roundtrip. Use it if
your interface and CPU support it without dropouts. Best feel for the
artist.
- 64 samples: comfortable safety margin, still under
~3 ms roundtrip on a typical 96 kHz interface. This is the
sweet spot for most live shows.
Rule of thumb for live voice: pick the smallest buffer
size that runs your busiest show without dropouts on your hardware.
Higher rate + small buffer beats lower rate + medium buffer for live
feel.
Test before the show. If you haven't run Clear Voice Live
on the channel count you'll have on stage, reduce the show's risk by
doing a 10-minute soundcheck with all channels armed. CPU-related
dropouts usually appear at the moment everything talks at once.
5. GUI Overview
The plug-in window is intentionally small and resizable. Everything you
can change is on screen, no hidden tabs or sub-pages.
The plug-in interface — three faders, two safety buttons, two engines, three meters, and the PRESETS button in the header (centred above the buttons, level with the trial/version text).
Header
- Title plate — click anywhere on it to open the about overlay (legal info, license attributions, version).
- PRESETS button — opens the preset menu: factory presets plus your own preset files. See section 9.
- Status LED + label — green & "ACTIVE" when audio is processed, off when bypassed. During trial it shows "TRIAL Xd" or "TRIAL mm:ss" countdown.
- Version — current plug-in version, top right.
6. The Three Faders
STRENGTH (0 … 10)
The master amount of denoise / dereverb processing. Fader at 0
means the plug-in is effectively bypassed (gain = 1 across all
bands). Fader at 10 means full processing.
Use this fader to dial in how aggressive Clear Voice Live is.
Below ~5 you'll usually have transparent denoising; above 7 the
intervention becomes audible to a careful listener.
SMOOTH (0.3 … 4.0)
The smoothing-time multiplier for gain decisions. Default 1.0.
- Lower values (0.3 – 0.7): faster gain changes, more responsive on
attacks but can introduce micro-pumping on dense material.
- Higher values (1.5 – 4.0): slower gain changes, very stable, but you
may lose ultra-fast transient detail (sibilants).
Set Smooth last. Get Strength and Dereverb where you want
them first. Touch Smooth only if you hear pumping (lower it) or if voice
feels unstable (raise it). For most live channels, the default 1.0 is
the right value and you never need to touch this fader.
DEREVERB (0 … 10)
Per-bin tail suppressor with voice-activity gating. Reduces the room tail
that follows each syllable, plus mid-tail reflections. At 0 the plug-in
is transparent; at 10 the voice sounds tight and close-mic'd even in a
room with 3 seconds of natural reverb.
Important — Dereverb only works in combination with Strength.
If Strength is at 0, Dereverb does nothing — the entire processing pipeline
is bypassed. To hear Dereverb's effect, raise Strength first.
Use Dereverb even when there's no audible reverb. A small
amount of Dereverb (1 – 3) strengthens the feedback prevention
by tightening the spectral gating around tonal resonances. We recommend
always running with a non-zero Dereverb in live situations, even on a
dry stage.
7. Engines: Soft & Hard
Two engines are selectable below the Bypass / Delta buttons:
| SOFT | HARD |
|---|
| CPU | Ultra-low — this is the engine behind the
"dozens of instances at a few percent CPU" figures. |
High — several times the CPU of SOFT per instance.
Not intended for many simultaneous instances. |
|---|
| Frequency resolution | Standard | Higher — finer-grained spectral analysis |
|---|
| Sound character | Gentler. Voice is barely touched even at high Strength. | Tighter, more aggressive. Higher noise reduction at the same Strength setting. |
|---|
| Best for | Most live channels. Default starting point. | Difficult rooms. Heavy bleed. Channels that need the heaviest help. |
|---|
| Channels per show | Many — run on every vocal mic. | Selective — use on the channels that need it most. |
|---|
Think of SOFT as the standard mode and HARD as the
specialist mode: HARD trades a lot of CPU for a more aggressive
spectral grip — it is not simply "more quality". On most
voices SOFT sounds just as good or better, at a fraction of the CPU. A
practical full-show setup: SOFT on every vocal channel, HARD only on the
one or two channels that genuinely need the heaviest cleanup.
Always start with SOFT. If after dialling in Strength
and Dereverb the cleanup is still not enough, switch to HARD and start
Strength back at ~30 %. You'll usually arrive at a more transparent
result than SOFT pushed all the way up.
If soundcheck time permits, try both engines on the channel.
Different voices respond differently — a deep male spoken voice may sound
cleaner under HARD, while a bright female lead may prefer SOFT. There is
no universal "right" choice; let your ears decide per channel.
Two safety buttons in the centre column. Both are locked by default
to prevent accidental clicks during a live show. Each button has a small
amber lock strip below it.
How the lock works
- Click on the lock strip below a button → toggles between
Locked (amber) and Armed (dark).
- Click on the button itself → toggles the action, but only if the
button is unlocked.
- The locked state persists across DAW sessions.
BYPASS
When ON, the plug-in is in audio bypass — the dry input passes straight
through. Useful for rapid A/B comparison ("how much better is this with
the plug-in on?").
DELTA
When ON, the plug-in outputs only the difference between the
processed and the dry signal. You hear what the plug-in is removing —
noise, room tail, feedback resonances.
Delta is a "nice to have" inspection tool. Never set values while
in Delta mode. Delta lets you hear what's being removed, but it
is not a tuning aid — the spectral character of the residual is not
representative of how aggressive the processing is. Set values with the
plug-in in normal mode, by ear, and use Delta only as a curiosity.
⚠ Warning — Delta Output Can Cause Feedback
Only enable DELTA when the channel is muted at the desk,
or when monitoring through closed-back headphones in solo.
Never
switch Delta on while the microphone is open through a monitor wedge,
in-ear pack, or PA — the residual signal from the AI gating contains
unpredictable spectral peaks that can drive the system into instant
feedback or sound horrendously to the audience. Delta is a studio /
diagnostic tool, not a live tool.
9. Presets
The PRESETS button in the header opens the preset menu.
It has two parts:
- FACTORY — eight built-in recipes distilled from our
measurement sweeps (also available in your DAW's own program list):
Default, Extreme Speech Denoise & Defeedback,
Hard but Gentle, Golden Speech,
Hard Music Debleed, Super Soft (Denoise only),
Soft Sweet Spot, Soft & Gentle.
- USER — your own preset files, loaded from
Documents\Clear Voice Live\Presets\
Saving and managing your own presets
- Save preset… stores the current settings (Engine,
Strength, Smooth, Dereverb) as a
.cvlpreset file in the
preset folder.
- Open presets folder opens that folder in
Explorer/Finder.
- The file name is the preset name. Rename a file to
rename the preset; copy files between machines to share presets. The
menu re-scans the folder every time you open it — files you drop in
appear immediately, no restart needed.
A .cvlpreset file is plain, readable text — you can view
and edit it in any text editor:
# Clear Voice Live Preset
version=1
engine_type=0 (0 = SOFT, 1 = HARD)
strength=100 (0–100, the fader shows this ÷10)
smooth_mul=1.64 (0.25–4.0)
dereverb_amount=0 (0–100, the fader shows this ÷10)
10. Status, Trial & License
The header shows one of these states at all times:
| Header text | Meaning |
|---|
| ● ACTIVE | Licensed, audio is being processed. |
| ● BYPASS | You pressed the BYPASS button. Click again to engage. |
| ● TRIAL Xd | Trial mode, X days remaining. |
| ● TRIAL mm:ss | Trial mode, less than an hour remaining. |
| ● EXPIRED | Trial period has ended. Audio is bypassed. |
| ● TAMPERED | Trial integrity check failed. Audio is bypassed. Contact support. |
Trial behaviour
A 21-day fully-functional trial begins on first plug-in instantiation. The
trial is bound to the machine you started it on. Renewing the trial
requires support assistance.
When the trial expires, the plug-in keeps loading and shows an "Expired"
overlay, but audio is bypassed cleanly until a license is activated.
11. Tips & Best Practices
- Close the plug-in window during the show. The GUI
(meters, animations) costs extra CPU; with the editor closed every
instance runs in its minimal audio-only path. Open windows only while
you're actually adjusting something.
- Always start with SOFT at 0 % and increase. Most
channels never need HARD. If SOFT at high Strength still isn't enough,
switch to HARD and restart at ~30 %.
- If soundcheck time permits, try both engines per channel.
Different voices respond differently to SOFT vs HARD. There's no
universal best choice.
- Set the order: Strength → Dereverb → Smooth. Dial in the
noise reduction first, the room tail second. Touch Smooth only if you
hear pumping or instability.
- Run a small Dereverb even on dry stages. 1 – 3 on the
Dereverb fader strengthens feedback prevention measurably.
- Lock the buttons before the show. Both Bypass and
Delta default to locked. If you've armed them during sound check,
lock them again before doors open.
- Never enable Delta on a live mic into a monitor or PA.
See section 8 — Delta can cause feedback. Use it muted or in
headphones only.
- One instance per channel. Don't chain multiple
instances of Clear Voice Live in series; the AI gating doesn't
compound usefully and you'll spend CPU twice.
- Insert before compressors and EQs. The plug-in
expects a relatively clean dynamic input. Heavy compression
before the plug-in muddies the noise floor and makes the AI
work harder.
- Conferences & panel discussions: one instance per
microphone always gives the highest quality — each channel gets its own
AI. But a very good option for large panels: route all microphones
through an auto-mixer (e.g. a Dugan-style mixer) into a single
group, and place one instance of Clear Voice Live on the
group. You save CPU, get consistent treatment across the
table, and the auto-mixer keeps unused mics out of the path before
they reach the plug-in.
12. Troubleshooting
The host doesn't show the plug-in at all
- Use the installer instead of copying files by hand. It
puts the bundle in the right place and installs the required Microsoft
runtime.
- If you must copy manually: copy the entire
Clear Voice Live.vst3 folder to
C:\Program Files\Common Files\VST3\ — that is the standard
VST3 location every host scans.
(...\Common Files\VST without the "3" is the old VST2
folder; most hosts also ignore
%localappdata%-based VST3 paths.)
- On a freshly installed Windows (including Windows 10/11 LTSC) the
Microsoft VC++ runtime may be missing — without it the
plug-in fails to load silently. Install it once from
aka.ms/vs/17/release/vc_redist.x64.exe
(official Microsoft link), then rescan plug-ins in your host. Recent
Clear Voice Live installers do this automatically.
- Make sure your host is 64-bit and actually scans VST3 plug-ins.
The plug-in is bypassed and shows "UNSUPPORTED SAMPLE RATE"
Your DAW or interface is running at a sample rate other than 48 / 96 / 192
kHz. Switch your project to one of the supported rates and reload the
plug-in. Most live setups use 48 or 96 kHz.
The plug-in is bypassed and shows "TRIAL EXPIRED"
Your 21-day trial has ended. Your audio passes through unprocessed
(clean bypass — nothing is muted). Purchase a license at
clearvoice.live and follow the
activation steps in section 10.
The plug-in is bypassed and shows "TAMPERED"
The trial integrity check failed. Contact
info@clearvoice.live for a reset
and include the machine fingerprint visible in the About overlay.
I hear crackles / dropouts
CPU is at the limit. Don't raise the buffer size — that
defeats the purpose of running a zero-latency live plug-in. Anything above
~64 samples becomes unworkable for live use.
Instead, check what else is competing for CPU on your machine. Open the
Windows Task Manager (Ctrl+Shift+Esc) or macOS Activity Monitor and
look for:
- Background syncing (cloud-storage clients, OneDrive, Dropbox, iCloud)
- Other audio software running in parallel
- Antivirus or backup scans
- Browser tabs with heavy JavaScript / streaming
- Windows Update or system maintenance tasks
Close anything you don't strictly need during the show. If CPU is still
borderline after that, switch some HARD-engine channels to SOFT.
The voice sounds nasal / hollow
Strength is too high. Reduce it until the voice character returns. If
the noise floor still bleeds through at lower Strength, try switching
from SOFT to HARD and re-dialling Strength to ~30 – 50 %.
I can't hear what Dereverb does
Dereverb requires Strength to be non-zero — it's part of the processing
pipeline that Strength enables. Set Strength first, then bring up
Dereverb.
13. Support
Questions, bug reports, license issues:
Email: info@clearvoice.live
Web: clearvoice.live
Open-source licenses for components used in this product are listed inside
the plug-in's About overlay and at
clearvoice.live/open-source.html.
End-User License Agreement: clearvoice.live/eula.html.
Privacy Policy: clearvoice.live/privacy.html.