177 MB

Roland JV-880 Emulator. Emulator of a famous 1U rack unit rompler made by Roland in 1992, as a VST3 and AU plugin. Based on NukeYKT’s SC55.

Visit Developer Site

VirtualJV.1.0.1  ( 177 MB )

364 MB / 1.44 / 1.64 GB

Professional DJ Preparation. ABYZOR Keys is a modern DJ library preparation tool built for harmonic mixing and intelligent playlist management.

Analyze your music library, detect musical keys and BPM, organize tracks automatically, build better DJ sets, and export your playlists with confidence.

🎵 Professional Key Detection
🥁 BPM Detection
🎼 Circle of Fifths
📚 Smart Music Library
⚡ Smart Collections
🎧 Playlist Preparation
📈 Library Statistics
💿 Windows & macOS

Features:
– Accurate key and BPM detection with multi-threaded audio analysis
– Intelligent playlist management and harmonic mixing tools
– Built-in Circle of Fifths visualization for harmonic mixing
– Smart Collections for organizing tracks and advanced library review
– Export to M3U, M3U8, CSV, and DJ software like Rekordbox, Serato, Traktor, Engine DJ
– Professional upgrade includes Smart Set Builder, Mix Coach, and USB preparation tools
– Modern dark interface designed for DJs with waveform visualization and audio preview

Visit Developer Site

Download  ( 1.6 GB )

208/211/207 MB

Altitude is a modern musical instrument that combines subtractive, sampling and FM synthesis, enabling you to produce a wide variety of complex sounds. Booming basses, provocative pads, effervescent effects, sonorous sequences: whatever your mix needs, it’ll only be a few clicks away.

Ready for work.
Altitude is a modern musical instrument that combines subtractive, sampling and FM synthesis, enabling you to produce a wide variety of complex sounds. Booming basses, provocative pads, effervescent effects, sonorous sequences: whatever your mix needs, it’ll only be a few clicks away.
The clear user interface serves as your guide through your auditory adventures. With its straightforward layout, exploring the deepest depths of the audio engine’s facilities is just as effortless as making your first patch. The design is alive with visual aids and reactive displays that make it a cinch to work on intricate patches.

Unlock your creativity.
Altitude is bundled with a large factory ROM, full of instantly usable samples that are perfect for layering. Fine tuning your patches’ timbre has never been easier. And with a carefully curated selection of over 300 built-in patches to demonstrate every corner of the audio engine’s potential, you’ll never be stuck for ideas again.
A few clicks in the patch browser here, and a quick browse of the sample picker there; Altitude makes the sound design process feel natural.

Altitude sports two 24-step sequencers both with builtin arpeggiators. The sequencers can be routed to different generators. Each sequencer has 5 tracks, one for each property of the steps. Every track has a customisable length, creating new combinations of parameters each time the tracks loop.
You can randomize individual tracks in the sequencer for endless inspiration. And you can keep the pitch track locked to a specific scale for more predictable results.

It all starts here.
Every instance of Altitude has 4 luscious virtual analog oscillators and 4 retro-inspired sample layers.
All the generators have numerous FM routing options, as well as a selection of unison modes: Super, Hyper and Modern.
And with the keyboard mapping features, you can configure exactly how the generators are assigned to incoming notes.

Shape, sculpt and scream!
The two included filters can run in a wide range of modes and algorithms. Perhaps you’d like a classic transistor ladder filter, or maybe you’d prefer the highly resonant Sour algorithm. Altitude’s got you covered.
The filters can be routed in any number of ways. Serial, parallel, split, inverted, or even something inbetween.

Altitude’s carefully designed user interface is full of polish and attention to detail. Many parameters have preview displays that show up as you modify the controls, and then automatically hide to reduce clutter. Hovering over a modulated parameter highlights the sources that are assigned to it. And there’s many more workflow-enhancing features just like these to discover.
Four interface skins are included with Altitude, and you’re able to load in custom skins if you want to completely adapt the plugin to match your creative space. The built-in skins scale from 100% to 300%, filling even the largest of monitors all while staying pixel perfect.
And with ultra-fast load times, the interface is always ready to channel your imagination.

17 high-quality effects.
Altitude comes loaded with 17 effect unit types, ranging from traditional favourites like the Phaser and Reverb, to the more extreme, such as the Bitcrusher and Two-Tap Delay.
You can place up to 3 units of each type in the 8 effect slots in any order you like.
The 8 effect slots are grouped into two chains of 4. The chains can be routed either serially or in parallel, allowing for a wide variety of creative options.

Altitude is packed with modulation options. There are 5 envelopes, 4 math modulators, 2 step modulators, 4 LFOs, in addition to a matrix with 21 slots.
The workflow of assigning and tweaking routings is quick and full of visual feedback. To make an assignment, simply click on the modulation slot and then the parameter to modulate. You can then adjust the modulation amount. In the LFOs, you can also switch between unipolar and bipolar routing using the adjacent toggle box.
Highly tuned performance.
Altitude is one of those plugins that you don’t need to feel guilty about when adding instances into a project.
It automatically detects the features on your CPU and selects the most optimal code, to ensure the best performance.
With native support for Intel processors over a decade old all the way up to the latest Apple Silicon processors, Altitude is happy to adapt to whatever you throw at it.

The latest technology.
Altitude leads the way to the future of audio plugins, built from the ground up for the new plugin format CLAP.
CLAP is based on an extensible architecture. If both your host and the plugin support a given extension, you’ll get access to its advanced features.
Altitude already supports many extensions, such as polyphonic modulation, preset discovery, parameter indication, remote controls and more.

Absolutely no copy protection.
Once you’ve bought a copy of Altitude, you can install it on as many computers you want. You’ll never need to connect it to the internet, and you’ll never need to worry about license keys.
It just works, straight out of the box, forever.
Product specifications:
4 virtual analog oscillators
4 sample layers
3 unison algorithms
15 FM sources
2 filters with 15 algorithms
5 envelopes (1 ADSR and 4 DAHDSRs)
2 step modulators
4 math modulators
4 LFOs
21 slot modulation matrix
4 macro dials (2 XY pads)
17 effect unit types
4 reverb algorithms
8 distortion algorithms
8 effect unit slots
3 effect chain routing modes
2 24-step sequencers and arpeggiators
39 modulation sources
510 modulation destinations (251 polyphonic)
265 MB of sample data
335 factory patches
4 factory skins, and custom skin template
MTS-ESP tuning support

System requirements.
5 GB of available drive space (exact size depends on the number of installed formats)
1 GB of RAM
64-bit Intel, AMD or Apple Silicon processor
Supported operating systems: Windows 7 64-bit or newer; macOS 10.14-13; X11/Linux with glibc 2.22 or newer
Supported host software: CLAP; Audio Unit (macOS); FL Studio native
For the standalone format: MIDI input and audio output devices

Visit Developer Site

Altitude.1.6.9.WIN  ( 232 MB )

Altitude.1.6.8.MAC  ( 236 MB )

Altitude.1.6.8.LINUX  ( 232 MB )

26 MB

A Windows/cross-platform VST3 + Standalone + Android 15 Port synthesizer plugin that wraps the bit-accurate VDX7 Yamaha DX7 emulation (real DX7 firmware ROM running on an emulated HD6303 CPU + emulated OPS/EGS sound chips) behind an original, Dexed-inspired editor GUI built in JUCE.

The DX7 firmware ROM and the factory voice ROMs ship inside the project as the raw binaries in Firmware/ (dx7.bin, voices.bin, example.ram), embedded into the plugin at build time. You do not need to supply any ROM yourself.

What it does
REAL SOUND: MIDI in → authentic DX7 FM audio out, generated by the emulated firmware and chips (not a DSP re-implementation).
REAL LIMITATIONS: The original sound engine only supports the keyboard range of the original hardware, and tweaking knobs while sound plays is not supported.
Original 6-operator editor: every voice parameter (per-operator envelopes, frequency, keyboard scaling, plus global algorithm / feedback / LFO / pitch envelope) is editable with rotary controls.
Live emulated LCD and voice-number LED read straight from the running firmware.
Factory ROM browser (ROM1A … ROM4B, 256 voices) plus an INIT voice.
FILE menu (SysEx import/export):
Import bank – load any standard 32-voice .syx bank (4104-byte bulk dump); its 32 voices appear as a USER bank in the browser.
Export current voice – save the current (edited) voice as a 163-byte single-voice .syx.
Export bank – save the current bank as a 4104-byte 32-voice .syx, with the current edited voice written into its slot.
Live edits / automation: turning a knob sends a lightweight DX7 parameter-change SysEx (the firmware boots with SYS INFO AVAIL enabled), so every new note instantly reflects the current knob/automation values with no patch-reselect glitch. Note: like real DX7 hardware, editing a parameter sends any currently-held note into its release – a single sustained note will not morph in place, but automating across a playing sequence is smooth.
Accessibility: every knob has a full descriptive title (e.g. “OP3 EG Rate 1 (attack)”) exposed to screen readers and shown as a tooltip, is keyboard-focusable and arrow-key manipulatable, and its whole tile is grabbable.
Patches are saved with the host session.

Changes to the Android Version over the VST3 version
The .vst3 version does not contain the FX section the android version includes. This is because I wanted the VST3 version to be a faithful port of the original (V)DX7 project, as the original did not have any effects either from what I know.

The Android version has an FX button in the top left, which switches to the FX viewer with Chorus, Delay, Phaser and Reverb Effects. These are direct copies of the modules from my synth Aquanode Modular. I added these as I thought the DX7 can sound rather dry without any effects, and you can’t easily add FX like you can in your Windows DAW for example.
You can add these effects on windows when you load my Aquanode Modular FX on a mixer track in your DAW.

This project is not affiliated with or endorsed by Yamaha. “DX7” is a trademark of Yamaha Corporation and is used here only descriptively.

Visit Developer Site

VirtualDX7  ( 26 MB )

9.15 MB

HTML-based instrument for extreme time stretching. LOOM is a browser instrument that stretches a sound in time to build textures, pads and drones. It can make a short sample last for minutes without changing its pitch. It runs as a single HTML file, works offline, and needs no installation.

How to use it
1. Load a sample. Click + sample. WAV, AIFF / AIF, MP3 and FLAC are supported.
2. Choose a region (optional). Drag on the source waveform to select part of the sound. This is useful, because a very large stretch of a long file produces very long output. Click once to set a start point; double-click to clear the selection.
3. Listen first. Press Audition to hear the whole sample, or only the selected region.
4. Set the controls (see below) and press Render. The work runs in the background with a progress bar, and you can Cancel.
5. Play and check. Press Play. You can click anywhere on the rendered waveform to start playback from that point — useful when you want to hear only the end of a long texture.
6. Export. Choose WAV or AIFF, choose the bit depth, and press Download.

The controls
– Stretch — how much longer the sound becomes (time only). The slider goes very far; the number box accepts even larger values.
– Pitch — transposition in semitones, fully separate from stretch. The range is very wide, several octaves up or down.
– Weave — how many grains overlap at each moment (1 to 16). Low values sound rough and pulsing, and broken at 1. Higher values sound smooth and glassy. The strongest change is between 1 and 4.
– Scatter — how much the phase is invented (0 to 100%). At 0 the real phase is kept, so the stretch stays coherent and attacks survive. At 100% it is the full texture effect.
– Drift — the reading position wanders. Each grain reads from a slightly different, moving point in the sample. At 0 it is off. Higher values make the texture breathe and change over time, and make the higher Weave values meaningful.
– Reverb and Size — an internal reverb (amount and decay length) added after the stretch.
– Gain — output level.
– Normalize — brings the level up to a safe maximum.
– Mono → stereo — from a mono sample, builds a wide stereo texture using two independent phase passes.
– Reverse — plays the stretched sound backwards.

Stretch, pitch, weave, scatter and drift are part of the engine, so a change needs a new Render. Reverb, size, gain, normalize and reverse are applied afterwards, so you hear them at once.

Visit Developer Site

LOOM  ( 9 MB )

94 MB

Phizmo is a transwave / wavetable synthesizer inspired by the Ensoniq Fizmo. It closely emulates its behaviour, but is not a 1:1 clone. Phizmo scans through multi‑cycle wavetables (“transwaves”) along editable evolution curves, with two oscillators per voice, a full modulation set, an arpeggiator, per-voice insert effects and global reverb, and a set of
real‑time performance macros.

Phizmo looks for its content in your Documents folder. Create this structure and copy the files in as described in the download.

Presets in Documents/Phizmo/Presets show up in the LOAD browser and are where SAVE writes to.
Wavetables in Documents/Phizmo/Wavetables appear in the per‑oscillator LOAD menu (the little LOAD button inside each wavetable viewer). Presets already embed the wavetables they use, so a preset will load its sound even if the loose .wav isn’t installed, but you need the Wavetables folder if you want to load tables into oscillators yourself. You can also simply drag‑and‑drop a .wav straight onto a wavetable viewer, same for presets.

Voice vs. Oscillator (the core concept)

Phizmo has two levels of sounds, and the panel is labelled to make the scope of every control clear.

Voice (4 of them – the EDIT / ON VOICE row, top‑left)
A Voice is a complete layer/patch with its own keyboard zone. You can stack or split up to four Voices across the keyboard to build a single
sound. Each Voice has its own LOW KEY / HIGH KEY range, its own level, and its own full parameter set. (Internally the code calls a Voice a “Sound.”)

Oscillator (2 per Voice – the EDIT / ON OSC row)
Inside every Voice are two oscillators, OSC 1 and OSC 2. Each has its own transwave, evolution curve, tuning, pan, level and wave settings. Ring modulation and the A/B mix combine them.

So the full layout is 4 Voices × 2 Oscillators = up to 8 Transwaves playing simultaneously.

Every section title shows what it affects: GLOBAL, PER VOICE, OR PER OSCILLATOR.

EDIT buttons (1–4 for voices, OSC 1 / OSC 2 for oscillators) choose what the panel is editing right now. They don’t change the sound, they change what the knobs point at. ON buttons switch a Voice or an Oscillator in or out of the sound.

Transwaves & evolution (how the sound moves)

Each oscillator plays a transwave: a wavetable made of many single‑cycle frames. Instead of a static waveform, Phizmo scans through the frames
over time, which is what gives it its living, morphing character besides the unique tonal character of the transwaves themselves.

The evolution editor (the big graph under each wavetable viewer) is the path the scan follows. Draw the curve, or use FLAT / RAMP / STEP presets. EVO sets how fast the scan moves; SCAN sets the direction/style (forward, backward, ping‑pong, one‑shot…). The moving dot on the curve is the live playhead. STEP mode jumps frame‑to‑frame.

Credits

The whole concept of Transwaves and the design idea belongs to Ensoniq. The transwave wavetables and the preset names as well.

The code, the DSP engine, the GUI, the preset/wavetable system and all the fixes along the way, was written by Claude Opus 4.8, the AI by Anthropic, working from my direction across a series of conversations.

Visit Developer Site

Aquanode.Phizmo  ( 123 MB )

259 MB

Fragile Violin started with a simple idea: to capture the quieter, more vulnerable side of the violin.
Not the big polished Hollywood string sound, but the small human details that often get lost in traditional sample libraries: the bow noise, the soft attacks, the uneven pressure, free bowing movement, and the feeling of a real player sitting close to the microphone.

I worked with multiple players to get too interesting articulations that are out of the norm.

Every sample we recorded are around 1:30 minutes long violin performances with multiple players feeding of off each other. This leads to a Sampler that has reel movement and is constantly changing, without ever hitting the looppoint. Every sample has 5 round robins to keep the sound interesting.

This free version gives you a small taste of a larger instrument built around fragile Violins, unusual articulations, and emotional imperfections.

The library is heavy reliant on the CC1 Modulation wheel, this is what brings the library to life.

Every Articulation has 3 different “Mixes”
Mix mode- normal mix,
Warp mode – all the samples are re-recorder through a broken Tascam 422 – 4 track.
Soma Mode – in this mode all the samples are procesed through the Soma Lyra FX chain plus extra analog filtering.

As a film composer, I often look for sounds that do not immediately scream “epic,” but instead sit close to the picture, breathe, and feel human, and have built-in mistakes, That is what I wanted this instrument to be. I hope you make something beautiful with it.

Visit Developer Site

Fragile.Violins.Free  ( 259 MB )

16.2 MB

A finite-difference time-domain physical modelling synthesiser.

1. Model
PartialString simulates the vibration of a plucked string by numerically solving the one-dimensional wave equation in real time using the finite-difference time-domain (FDTD) method. The string is represented as a list of displacements in memory. A lower note requires a longer string, which needs more points to represent it. You can then excite the string at any point along its length, and the FDTD scheme will compute how that change in displacement propagates. A pickup measures the displacement at a chosen location, and this generates the audio output. 10-voice polyphony is supported (up to 10 simultaneous strings), but this is dynamically adjusted based on your computer’s capability and the lengths of the string being simulated.

2. Controls
You can set the position and shape of the excitation, both velocity-sensitive. An LFO can sweep the pickup position, up to audio rate for AM-like sidebands. The string’s decay time is adjustable, with or without dampers applied. The dampers can behave like a piano’s, closing when a note ends. Combined with a long undamped decay, this makes pad-like sounds. Reducing the simulation accuracy generates unrealistic, soft and inharmonic sounds. Accuracy can be set constant for all notes, or proportional to string length (which can help prevent CPU overload on less-capable machines).

3. Performance
Simultaneously simulating many low notes (long strings) accurately is quite computationally intensive. On less capable computers, PartialString will automatically limit the number of sounding voices to prevent CPU overload, and this is indicated on the UI. You may notice that the synth behaves monophonically in low registers.

There is an option to prevent too-low notes (those estimated to cause overload on your machine even when played monophonically) from sounding, or to automatically pitch them up an octave or two to maintain performance. This option only takes effect when the synth is being used in real-time (i.e. not when your DAW is bouncing or freezing tracks).

On Apple Silicon Macs (M-series), running an Intel-based DAW forces the plugin to run through the Rosetta 2 translation layer, which can significantly reduce CPU performance. For the best results, ensure you install and run an Apple Silicon native (or ‘Universal’) version of your DAW.

Visit Developer Site

PartialString.1.0.3  (16  MB )

36 MB

Turn any folder of samples into a playable instrument.
Drop a folder of samples and Pathfinder scatters them into a living field of light. Draw a path to play it, pull a sphere apart into tempo-synced grains, and shape the whole thing with effects.

Four moves. That’s the whole instrument.
No menus to learn. You drop, draw, explode, and shape — and the field responds in real time.

Drop in a folder
Drag any folder of samples straight onto the plugin — it analyzes each one and scatters them into a 3D field. Brighter, closer spheres are the sounds it likes best for playing.

Draw a path to play
Trace a line through the field and Pathfinder plays every sphere it crosses, in order. The shape of your line becomes the rhythm and melody of the phrase.

Explode a sphere
Grab a single sound and blow it apart into tempo-synced grains. One sample becomes a swarm you can re-scatter, re-pitch, and play back as a texture.

Shape it with FX
Drag effects into any order — reverb, degrade, flair, delay and a granular engine — and reorder the chain live to sculpt the field into a finished sound.

A granular sampler that feels like an instrument.
Your samples, mapped
Nothing is baked in. Feed it your own drum kits, field recordings, or one-shots and it builds a fresh field from whatever you give it.

Drawn, not programmed
Paths are performances. Draw fast or slow, jagged or smooth — the gesture is the sequence, so no two takes are the same.

Tempo-synced grains
The granular engine locks to your DAW. Explode a sound and every grain lands on the grid, from tight stutters to slow, drifting clouds.

// the toolkit
Everything inside.
A whole sampler lives under the map — scan, draw, sequence, edit and process, all in one instrument.

Particle map
Every sample becomes a sphere, auto-sorted into kick, snare, hat, perc, melodic, texture, vox and FX.

Category detection means you find the right sound by where it sits — not by scrubbing filenames.
Draw the path
Play any route from MIDI, tempo-locked and pitch-corrected to your project key and scale.

Key detection lets samples from totally different sources play in tune together — no manual retuning.
Layer mode
One key fires your whole captured group as an instant stack.

Explode
Select a sample and slice it into tempo-synced chunks scattered back onto the map.

A fast way to turn one hit into a whole palette of rhythmic fragments.
Step sequencer
Per-lane sounds, with swing and nudge.

Deep sound editor
Trim, envelope, pitch and volume per sound, plus a chromatic pitch-tracked keyboard for any sphere.

Five-effect chain
Reverb, Degrade, Flair, Delay and a tempo-synced Granular engine — drag to reorder, with full parameter pages.

Presets that remember everything
Paths, sequencer, effects, categories and settings, all saved together.

MIDI CC learn · scalable interface · built-in walkthrough

// techniques
Things to try.
No two folders make the same map. No two paths make the same instrument. A few places to start:

In-key arps
Draw a path across only your melodic spheres for evolving arpeggios that stay in your song’s scale.

Granular textures
Explode a vocal or pad, then draw a new path through the fragments for moving, atmospheric fills.

Kits from chaos
Lasso your kick, snare and hat spheres and build a drum pattern in the step sequencer from found sounds.

Instant chords
Capture a melodic group in Layer mode to turn any folder into a one-key chord instrument.

Visit Developer Site

Pathfinder.1.0.4.WIN  ( 8 MB )

Pathfinder.1.0.4.MAC  ( 2 MB )

23.6 MB

Four-layer one-shot sampler for AU, VST3 and standalone use. One MIDI note normally triggers all four layers, or one non-repeating random layer in Round Robin mode. DICE selects samples from their source folders and randomises volume, pitch, start point and delay.

Features:
– Four drag-and-drop sample slots with waveform display and previous/next folder navigation
– Smooth proportional interface resizing from 60% to 200%, plus persistent scale presets
– Twenty persistent colour schemes with dark, light and colourful variations
– Optional Glitch Mode with per-note inverted colour flashes, regenerated Zalgo text, waveform artefacts, randomized delays, stutters, crackles and sample-hold glitches
– Refined arc-style controls with parameter units
– Per-slot volume, whole-semitone pitch with Shift fine mode (0.1 st), draggable start point, positive shift (0–250 ms), and fade-end controls
– Smooth real-time playheads and visual start, shift, and fade editing on every waveform
– Hover-highlighted Start and Fade markers
– Non-crossing Start/Fade markers, draggable range selection, and mouse-wheel fade editing
– Option/Alt + mouse wheel adjusts fade-in and fade-out curve shape
– Mouse-wheel slot volume editing with matching waveform amplitude scaling
– MIDI note-on triggering
– Selectable 1–16 voice polyphony
– Hold-to-repeat Burst mode with adjustable 1–30 Hz rate
– Per-trigger parameter evolution (PTE) with a real-time-safe audio path
– Optional chromatic MIDI keytracking relative to note 60
– Random Round Robin mode plays one loaded slot per trigger without immediate slot repeats
– Per-slot LOCK controls to exclude layers from DICE and PTE
– Per-slot random-sample and clear controls
– Separate DICE, SAMPLES and PARAMS actions
– BACK recalls the previous generated combination
– Four global randomisation-depth controls
– Real-time-safe master volume control
– DAW state restoration for parameters and sample paths
– Last-used sample folder is remembered globally; DICE can populate empty slots from it
– Once-daily background check for newer GitHub releases with an in-plug-in notification

Visit Developer Site

Sample.Dicer.1.5.2  ( 24 MB )