80 MB

Turn a single note into a living soundscape. Voyage Voyage is a freeware Shimmer Reverb built around an analog BBD chip delay emulation – ethereal pitch-shifted tails meet dark, grainy analog repeats that slowly dissolve into the mix. Push it further and it becomes an instrument of its own: an infinite drone and texture generator for ambient beds, cinematic tension and evolving sound design, using its self-oscillation and resonator capabilities.

A Shimmer Reverb and a Drone Machine, designed for the KVR Developer Challenge 2026, available for Windows, macOS and Linux with VST3/AU/CLAP formats.

Main features:

classic decay, tone, mix, input, output, low cut and high cut controls, as expected in a reverb plug-in.
Shimmer amount & pitch control, with an additional delay time control featuring the BBD chip emulation.
innovative Exciter & Chaos parameters, useful when Voyage Voyage is in self-oscillation / resonator / generation mode
more than 50 presets in 6 categories: delay, reverb, spacecraft, submarine, textures, and miscellaneous.
as usual with Musical Entropy products, resizing capabilities, integrated help, extra accessibility functionalities, embedded tutorial, mix lock control, easter eggs etc.

Visit Developer Site

VV.1.0.1  ( 81 MB )

10 MB

an ambient, spatial reverb — built on two layers, not one
A wide, continuous stereo wash — the sky — paints long, open tails with movement in them. Over it, a set of small binaural events — the stars — sit at real directions around your head. As you play, the stars answer your notes and flash exactly where you hear them; a spark on the right really is on your right. They can reply at once or hold back, free or locked to the host tempo, so the answers fall into the gaps between your phrases.

Two-layer engine — a dense stereo sky plus binaural 3D stars, decoupled from the tail
Stars that react to your playing, with a tempo-synced or free response delay
Eight factory presets, each a distinct space — Overcast, Clear, Sky Opening, Cirrus, Aurora, Storm, Vocal Sky, Cinematic Overcast
A/B compare, your own saved presets and favorites, and a freely resizable window
Headphones mode for full binaural placement; speaker mode for a wide, mono-safe fold
macOS VST3 & Audio Unit, Windows VST3 — free, with no time limit

Features
Two-layer engine: a dense stereo reverb bed plus binaural 3D events (KEMAR HRTF), decoupled from the tail.
Stars that react to your playing and answer at the exact direction you hear them.
Tempo-synced or free response delay, from instant to three seconds or 1–8 quarter notes.
Nine controls in three groups (SKY / STARS / OUT), each mapped to one audible result.
Eight factory presets, each a distinct space: Overcast, Clear, Sky Opening, Cirrus, Aurora, Storm, Vocal Sky, Cinematic Overcast.
Live “sky view” that shows the sound honestly, and slim output metering in the header.
Headphones mode for full binaural placement; speaker mode for a wide, mono-safe fold.
A/B compare to hold two settings side by side and copy one onto the other.
Save your own presets and mark favorites for quick recall.
Resizable interface — scale the window to your screen.
Mix lock to keep your wet level steady across presets.
Illustrated user manual included, reachable from the plugin’s menu.
MacOS 12+ universal (Intel + Apple Silicon); Windows 10/11 64-bit (also runs on Windows-on-ARM inside x64 hosts). Linux on the roadmap; usable now via yabridge.
Free, for any music or audio work, commercial or not, with no time limit.

Visit Developer Site

SkyVerb.1.0.22  ( 10 MB )

9.5 MB

Halide — Granular Smear & Spectral Wash
A grain cloud that feeds its own spectral ghost — dissolve the actual sequence of grains, not just the source.

Halide runs a 256-voice granular cloud into a Paulstretch-technique spectral smear — in series, not parallel, so SMEAR doesn’t add a second layered effect, it dissolves the grain cloud’s own output into slowly evolving, phase-randomized spectral fog. Feed it anything: a chord blooms into an evolving pad, a vocal line sings alongside its own harmony, a drum loop dissolves into an ambient wash that keeps breathing long after the source stops. Everything — density, size, shape, pitch, feedback, motion — shapes the fog, because the fog is made of the cloud.

Core Signal Path
256-Voice Grain Cloud — density and size fully decoupled, a bipolar SHAPE control running rexpodec Gaussian expodec (swell, fog, or pluck character), and a single SEED macro that scales timing/pitch/pan jitter together as one “character” axis.
Serial Spectral Smear — the grain cloud’s own output feeds a from-scratch Paulstretch-style engine (phase-randomized FFT resynthesis, pitch-locked crossfades on every re-anchor) — stretch 1–100×, selectable FFT window from watery-small to choir-huge.
Musical Pitch Mode — flip PITCH into an interval/chord selector (14 entries: intervals, triads, sus4, 7ths, and a bright SHIMMER stack) and every grain randomly picks a chord tone, so the cloud sings a harmony around whatever you feed it — no key entry, the source is always the root.
Feedback Through the Whole Chain — repeats re-pass through the grain engine and the smear both, so FEEDBACK builds shimmer (PITCH +12) or murk (PITCH −12) with each pass, not just a repeating delay.
Freeze + Motion — freeze the buffer and let MOTION auto-scrub through it for infinite, hands-free evolving pads from a single held instant.
Built to Watch, Not Just Hear
A live buffer waveform display shows the write head, the TIME playhead, and every actual grain spawn as a blooming, fading ember — amber cooling to redscale over each grain’s lifetime. It’s not a generic meter; it’s a direct readout of the cloud actually happening in real time, including exactly what SMEAR is currently dissolving.

Highlights

Ducked, equal-power Mix — the fog displaces the dry signal rather than stacking on top of it, so both extremes (bit-perfect dry, full wash) stay pure.
Loudness-matched wet path — grain window energy and a slow servo both compensate automatically, so MIX behaves predictably at any SHAPE or SMEAR setting instead of getting quieter as the texture changes.
Pitch-locked spectral wraps — the smear re-anchors without the pitch clash that plain crossfading between differently-pitched material would cause.
BPM-syncable MOTION and FREEZE, for scrubbing and rhythmic texture work locked to the host tempo.
What It’s For
Ambient pads and evolving washes from static sources, granular vocal and lead harmonization, shimmer/murk feedback textures, freeze-and-scrub sound design, and any material that needs to dissolve into something that keeps moving after the input stops.

Halide is donation-ware — free to use, no strings attached. If it earns a place in your workflow, a tip is always appreciated: buymeacoffee.com/brittnell
Format: VST3 (Windows & macOS) | Current version: 1.0.1.

Visit Developer Site

Halide.1.0.1  ( 10 MB )

11 MB

Scintillate is a Spectral Reverb Sparkle Engine.

Sparkles are generated within the spectrum of a conventional reverb.

At different spawn and decay rates, these sparkles can sprinkle fairy dust on your instruments, turn them into a fluttering wall of sound, and even create trippy reverse effects.

Reducing the density lets only the loudest sparkles through, making for color-ful spectral gate effects.

DSP by Signalsmith | Everything else by Sweet Audio

{See video at top of page}

This is the first release by Sweet Audio, and also serves as a technical test of our in-house plugin framework before launching bigger plugins later in 2026.

Scintillate and our other plugins are written natively as CLAP plugins, and made compatible with VST3 and AU using the open-source clap-wrapper. If your DAW supports CLAP, prefer to use it to benefit from native support for parameter modulation and integration into the host’s undo/redo stack.

About Sweet Audio
It’s Sweet Audio’s declared goal to build a customer-friendly and sustainable audio plugin company:

We won’t play the sale game to manipulate you into buying our products – all of our paid plugins will have a constant and actually reasonable price. We say no to Black Friday.
We also don’t do marketing spam – if you sign up to our mailing list, you get exactly one e-mail for every plugin we launch in the future. Visit sweet-audio.net to subscribe.
Sweet Audio is founded by Marius Metzger, the developer behind CrispyTuner. At the time, CrispyTuner was a one-off project that I ultimately sold to Plugin Alliance, who now distribute it as bx_crispytuner.

Visit Developer Site

Scintillate.1.0.0  ( 11 MB )

11.5 MB

GrainBrain — Stereo 4-Band Multiband Granular Processor
Split it. Freeze it. Scatter it. Bleed it into the next band.
Splits your signal into independent frequency bands, then shreds each one into grains — its own LFO, spectrum view, and gate per band. The deepest knob-forest in the XNULLX family, built for players who want their hands on everything at once.

Built for texture
INDEPENDENT BAND STRIP
Split the signal across multiple bands, each with its own granular engine — sculpt highs into shimmer while keeping the low end intact.

PER-BAND LFO MODULATION
Every band carries its own modulation source, so grain density, size, and position can drift independently across the spectrum.

LIVE SPECTRUM VIEW
Watch the band split and grain activity in real time instead of guessing at crossover points.

GATING PER BAND
Keep grains from spawning on silence — each band’s gate reacts to its own signal, not the full mix.

Core Signal Path
4-Band Split — draggable bandpass filters right on the spectrum analyzer; grab a band and move it, narrow it, or disable it entirely.
8-Voice Granular Engine (per band) — grain size, scatter, pitch + pitch spread, drive/saturation, and wet/dry mix, independently on every band. Reverse and Freeze modes included.
Cross-Band Bleed — each band’s output can feed into the next (Low LoMid HiMid High, wrapping back to Low), for resonant, self-reinforcing textures that a simple per-band effect can’t produce.
Stutter Gate (per band) — a BPM-synced trapezoid envelope with attack, release, and phase offset, for rhythmic amplitude patterns that lock to the host tempo.
7-Target LFO Matrix (per band) — modulate Center, Width, Mix, Pitch, Scatter, Size, and Gate Phase simultaneously, each with its own depth and phase.
Built to Watch, Not Just Hear
The real-time spectrum analyzer isn’t decoration — band regions animate live under LFO modulation, so you can see a Center sweep or a Width pulse happening in sync with the audio. Drag directly on the analyzer to reposition a band by ear and eye.

Highlights
Ducked, equal-power Master Mix — dry and wet displace each other instead of just stacking, so blending in granular texture never gets muddy.
Global In and Out trim sit alongside Mix in the bottom bar — set levels in one place without hunting through four bands of per-band Gain knobs.
Collapsible Spectrum / Gate / LFO sections, remembered per preset, so the window stays exactly as compact or as deep as the patch needs.
Full MIDI Learn on every parameter.
What It’s For
Multiband texture design, granular pads and drones, rhythmic stutter/gate effects synced to tempo, spectral freeze composition, glitch and sound-design work, and subtle “make it feel alive” granular blending underneath a mix — GrainBrain covers the gap between a delicate texture enhancer and a full sound-design instrument.

GrainBrain is donation-ware — free to use, no strings attached. If it earns a place in your workflow, a tip is always appreciated: buymeacoffee.com/brittnell
Format: VST3 (Windows & macOS) | Current version: 1.0.1.

Visit Developer Site

GrainBrain.1.0.1  ( 11 MB )

332 MB

Modular Filter/Delay. ShinRonin is a ground-up rebuild of Ronin, one of the earliest plugins we ever shipped, first released in 2003. The original paired two delay lines with a set of morphing filters and two fully patchable routing matrices, and it earned a cult following for the strange feedback networks and self-oscillating textures you could build inside it. Twenty years on, we have rebuilt it for modern systems, kept the architecture that made it worth remembering, and released it for free.

Two independent delay lines, each with tempo sync, reverse, looping, and feedback, feed a pair of morphing multimode filters that sweep continuously from notch through lowpass, bandpass, and highpass. There is saturation, two LFOs, and an envelope follower. Everything connects through two on-board matrices: a signal matrix that routes any audio source to any destination, and a modulation matrix that wires the LFOs, envelope, and incoming MIDI to nearly any parameter. It is happy doing clean stereo delays and gentle filtering, but the real fun starts when you patch a feedback loop and let it misbehave.

Visit Developer Site

ShinRonin.1.0.0  ( 333 MB )

37 MB

Tesseract Echo is a free, deeply tweakable Multi-Tap Magnetic Tape Delay and FDN Reverb.

Engineered for zero artifacts and maximum character, Tesseract Echo relies on Hermite 4-point interpolation, constant-power panning, and dual-stage diffusion to deliver pristine, wide, and deep spatial effects.

Under the hood:

4-Head Tape Engine: Independent level and pan controls per head, with fractional spacing algorithms.

Non-Linear Tape Filter: 1 pole Highpass and Lowpass filters, and an analog-style Drive stage inside the feedback loop with automatic makeup gain.

Advanced Modulation: Push into audio-rate modulation or use the “Tape” (Wow & Flutter) and “Stepped Random” LFO shapes for mechanical degradation.

Karplus-Strong Engine: Tune the delay time to incoming MIDI notes to synthesize physical strings and tuned resonances.

FDN Reverb & Ducker: A pristine Feedback Delay Network reverb with selectable routing, plus a built-in sidechain compressor to duck the wet signal against the dry input.

Download the plugin for free in multiple formats for Windows, macOS, and Linux.

⚠️ A quick note on donations (Thank you!)
This plugin is strictly 100% free! If you’d like to support my work and fuel the development of future plugins, I am incredibly grateful.
However, if you type a donation amount here at checkout, the platform takes a heavy cut in “shop commissions”.

To avoid these extra fees, please download the plugin for free (just type ‘0’), and consider dropping a tip directly into the Tip Jar / Goal on my main Ko-fi homepage instead! That way, your generous support goes entirely toward the DSP journey.

Visit Developer Site

Tesseract.Echo.1.1  ( 38 MB )

37 MB

YS – Roomer Delay_ An audio plugin for mixing. The plugin creates a delay effect. Three parallel delays can add a long echo effect or simulate a small room using short delays. Functionality: 1_Manual delay adjustment and linked to the project tempo. 2_High- and low-cut filters. 3_Panning of each channel. 4_Reverb added to the delay. 5_Three modes: 1: Mono (repeats are located according to the panorama as in the source) 2: Haas (a delay (1-100 ms) is added to one of the channels creating a pseudo-stereo effect) 3: Ping-Pong (alternating delay of the left and right channels) 6_ Channel phase inversion. 7_Switch to Mid-Side mode (For stereo sources or in Haas and Ping-Pong mode) 8_ LR-RL switching the direction of Ping-Pong and delay for the Haas effect. The plugin has three versions, two with different designs and sizes and one version with one delay module.

The plugin is made using Synthedit. (A universal VST3 plugin can run on both macOS and Windows. SynthEdit now releases universal VST3 plugins. This means you can offer the same download to both macOS and Windows users. It will “just work.” SynthEdit’s universal VST3 plugins support both Intel-based Macs and Apple Silicon (ARM)-based computers, such as the M1, M2, and later models.)

Visit Developer Site

YS_-_Rommer-Delay__v.1.0.2  ( 37 MB )

17.5 MB

Ambience is a high-quality, open-source algorithmic reverb VST3 plugin built on a 16-channel Feedback Delay Network (FDN) architecture. Designed with professional audio standards in mind, it delivers rich, natural-sounding reverberation ranging from intimate studio rooms to vast concert halls and beyond — with the precision and stability demanded by real-world production environments.

Schroeder, Moorer, Jot — for sixty years, the foundations of commercial reverb have barely moved. The differences are presets and UI.

Ambience reimplements that foundation from primary academic sources — FWHT feedback matrices, nearest-prime delay allocation, Välimäki-Liski 10-band GEQ, ISM-based early reflections. Research-paper DSP, end to end.

A Fast Walsh-Hadamard Transform forms the feedback matrix, delivering dense, colorless diffusion. The 16 delay lines are tuned to mutually coprime primes on a logarithmic scale, eliminating comb-filter artifacts at any room size.

ROOM1 / ROOM2 / HALL1 / HALL2 / PLATE / SPRING / GOLDFOIL — each algorithm uses a distinct topological routing, Allpass gain structure, and early-reflection pattern. Seven completely different reverb characters.

Each FDN channel hosts a 10-band biquad GEQ cascade. The plugin solves a Weighted Least Squares system in real time to follow target RT60 curves with surgical accuracy across the full audible band.

Early reflections are generated via the Image Source Method, tuned per algorithm for perceptually accurate pre-echo with full stereo imaging control.

Each of the 16 FDN channels is modulated by an independent bandlimited-noise LFO, initialized via a Golden-Ratio Weyl sequence to guarantee mutually asynchronous, non-periodic motion.

Anti-Derivative Anti-Aliasing in four flavors — Warm (Vicanek), Tape (Padé rational), Tube (asymmetric ADAA), Hard (ADAA clip). Texture without aliasing artifacts.

One PRO button unlocks per-band RT60 multipliers from 31 Hz to 16 kHz, Low / Mid / High Tilt EQ, and saturation type selection.
The live RT60 graph overlays the effective curve (orange — including HF Damping & LF Absorption) on top of the preset reference (gray). Mix by looking, not just by listening.

Every world-class hall, studio, plate and spring you’d wish for — modeled and included. No DLC. No upgrades.

Visit Developer Site

Ambience.1.0.1  ( 17 MB )

4.50 MB

Authentic 80s rack-style digital reverb. 100% wet output with vintage 12-bit character and Schroeder-style diffusion.

100% Wet Reverb
Pure reverb signal output – no dry signal in the mix. Perfect for:
– Parallel processing with dry signal in your DAW
– Creative sound design
– Authentic vintage rack reverb simulation

80s Rack-Style Character
– 24 kHz Internal Sample Rate – Authentic vintage digital reverb with band-limited frequency response
– 12-bit Quantization – Classic gritty digital character
– TPDF Dither – Smooth quantization
– Stereo Crosstalk – Authentic L/R channel bleed simulating analog circuitry
– Early Reflections – Program-dependent multitapped delay lines for realistic room ambience
– HF Damping – Mode-dependent high-frequency rolloff

Schroeder-Style Algorithm
Parallel comb filters with damping followed by series allpass diffusion.

Visit Developer Site

WetReverb-v1.0.1  ( 4 MB )