39/60 MB

Audio Plugin for Audio to MIDI transcription using deep learning.
NeuralNote is the audio plugin that brings state-of-the-art audio-to-MIDI transcription into your favorite Digital Audio Workstation.

What’s new in v2
A new, much more capable transcription model. v2 replaces Spotify’s Basic Pitch with MuScriptor, from Kyutai and Mirelo (paper, blog post, checkpoints). It is a transformer with 103M to 1.4B parameters, compared with fewer than 17K for Basic Pitch.
Multi-instrument transcription. Transcribe full mixes, not just one instrument at a time, with notes grouped by instrument.
A built-in multi-instrument synth. Listen to every transcribed instrument, with per-instrument gain, mute and solo.
A new, modern UI.
Still fully local. Transcription runs on your machine, on the GPU or the CPU, and your audio never leaves it. NeuralNote only goes online to download models and to check for updates.

Install
Download the latest release for your platform from the releases page.

macOS (Apple Silicon): NeuralNote_Installer_Mac_arm64.pkg installs the Standalone app, VST3 and AU. It is signed and notarized. An installer for Intel Macs will come later. In the meantime, you can build from source.
Windows (x64): NeuralNote_Installer_Windows_x64.exe installs the Standalone app and VST3. It is not code signed, so Windows may warn you before running it for the first time.
Linux: prebuilt binaries are coming soon. In the meantime, you can build from source.
The transcription model is not included. Download it from within NeuralNote the first time you open it (see Models and performance).

Usage
NeuralNote is a simple AudioFX plugin (VST®3/AU/Standalone app) that you apply to the track you want to transcribe.

Gather some audio:
Click record. This works both when recording live and when playing the track in your DAW.
Or drop an audio file on the plugin (.wav, .aiff, .flac, .mp3 and .ogg (vorbis) are supported).
Select the instruments to transcribe in the left panel, or select Automatic to let the model detect them.
Transcriptions tend to be better when the model is given the correct set of instruments.
Click Transcribe. The transcription runs in the background, and the MIDI fills into the piano roll as it is decoded. A progress indicator and a cancel button are shown while it runs.
Click play to listen to the result without waiting for the transcription to finish. Only the part decoded so far plays.
Adjust the mix between the source audio and the synthesized transcription.
Adjust the level of each instrument.
When you’re happy with the result, drag and drop the MIDI from the plugin onto a MIDI track, or save the MIDI file to your computer.
Models and performance
MuScriptor comes in three sizes, small, medium and large, which trade speed for quality. NeuralNote downloads them as GGUF files from DamRsn/muscriptor-gguf on Hugging Face. When no model is installed, all three sizes are available to download. The Model button in the top bar lets you switch between installed models, download other sizes, and open the folder where they are stored.

OS Models folder
macOS ~/Library/NeuralNote/models
Windows %APPDATA%\NeuralNote\models
Linux ~/.config/NeuralNote/models
The downloads are about 210 MB for small, 620 MB for medium and 2.7 GB for large.

You can also put a model file in that folder by hand. Use a file from v1/ of the HF repo, unchanged and with its original name (e.g. muscriptor-medium-f16.gguf). NeuralNote only picks up files whose name and size match the ones it downloads.

You can remove a model by deleting it from this folder, and re-download it at any time.

Hardware. Transcription speed depends mostly on the model size and on your hardware. The GPU is used when available, through Metal on macOS and Vulkan on Windows and Linux. A GPU is strongly recommended for the medium and large models. Settings > Compute device picks the device: Auto (the default, which names the device it chose), a specific GPU, or the CPU. The choice applies from the next transcription, and GPUs added or removed later are listed after restarting NeuralNote (or your DAW). Only a few GPUs have been tested so far. If transcription fails, gives wrong results or is unexpectedly slow on your machine, please open an issue, either here or in muscriptor.cpp if the problem is in the engine itself. Please include your OS, GPU and model size.

Approximate real-time factors measured on an Apple M1 Pro (above 1× means faster than real time):

Size GPU (Metal) CPU
small ~3.5× ~2×
medium ~1.5× ~0.7×
large ~0.5× not recommended

Visit Developer Site

NeuralNote.2.0.0  ( 103 MB )

10.6 MB

Call & Response MIDI is a generative composition instrument that replaces robotic randomness with musical dialogue. Built around the timeless musical principle of call and response, it generates interlocking melodic phrases, hypnotic acid basslines, and conversational hooks in seconds.

Whether you need a captivating lead, evolving counterpoint, or an instant burst of inspiration to break beat block, Call & Response MIDI gives you immediate musical results with complete tactile control.

Core Features

Conversational Dialogue Engine: Two interdependent lanes work in tandem—the Call introduces the musical motif, and the Response answers, inverts, echoes, or resolves it.
40 Musical Styles & 38 Response Strategies: Spans Acid, Techno, House, Synthwave, Neo-Soul, Ambient, and more with specialized algorithmic answers (Mirror, Invert, Retrograde, Question-Answer, Syncopation, Octave Leap).
35 Musical Scales with Dynamic Transposition: Switch between Diatonic modes, Pentatonics, Blues, and Exotic scales (Hirajoshi, Insen, Hungarian Minor, Byzantine). Changing scales or keys automatically transposes active notes while preserving musical intervals and melodic contours.
Interactive 2D Piano Roll:
Drag notes vertically to transpose pitch with instant audio auditioning.
Slide notes horizontally across step columns without scrambling the phrase.
Double-click to create or delete notes; drag right-edge handles to adjust gate length.
Built-in 303 Analog-Modelled Synth: Preview ideas instantly with an integrated acid synth featuring resonant low-pass filtering, envelope decay, accents, and slides—zero external routing required.
8 Instant Preset Slots (A–H): Save current states with Cmd+Click (or Ctrl+Click) and reload with a Double-Click.
Vector Parameter Locks: Click the padlock icons next to Style, Strategy, Scale, or Motif to freeze your favorite settings while randomizing the rest.
3-Way Split Views: Switch effortlessly between BOTH (dual stacked step grids + combined piano roll), CALL, and RESP.
Drag-and-Drop MIDI Export: Export generated phrases directly into your DAW timeline across 4 modes: Combined, Call Only, Response Only, or Split MIDI Channels (1 & 2).

Heritage & Attributions

Built upon the foundational algorithmic concepts originally developed by Nordl3 and popularized by SideBrain’s renowned Max for Live device, Viral Samples has engineered Call & Response MIDI into a native, high-performance C++/JUCE instrument for both macOS and Windows.

Visit Developer Site

Call.Response.MIDI.1.0.5  ( 11 MB )

50 MB

Harmony Lab Audio has released Harmony Pro, a free harmony assistant designed for musicians who want to explore chord progressions and alternative harmonic directions.

First, let’s get the GUI out of the way. It looks very vibe-coded, and since we obviously won’t be living in a world without vibe-coded plugins anytime soon, I hope at least these GUIs will improve. Just my two cents.

If we disregard the looks, Harmony Pro definitely has useful features. Its harmonic engine analyzes the current musical context and proposes chords that can continue or redirect a progression.

It’s also worth noting that this is not a prompt-based song generator. Instead, it suggests harmonically related options and leaves you to listen, compare, edit voicings, and assemble the result.

The developer is also explicit that no generative AI is involved inside the engine. The website sums it up as “No AI in this app. Just harmonic intelligence.”

To get started, you choose a tonic, a mode, and a style, then add or play chords and audition the suggestions. You can play from a MIDI keyboard and drag ideas straight into your DAW.

There are 13 style profiles, including Pop, Rock, Jazz, Neo Soul, Gospel, Lo-Fi Hip-Hop, Cinematic, and Progressive Metal, and the mode menu covers everything from major and minor to Dorian, Phrygian dominant, Hirajoshi, and whole tone.
What I liked is how deeply you can shape the suggestions. The engine has four presets (Balanced, Conservative, Exploratory, and Jazz Expert), plus controls for Richness, Boldness, Modal Character, Continuity, Exploration, and Style Fit.

You can also steer the direction toward Stable, Tension, or Resolution, with Color and Surprise as character choices

Visit Developer Site

Harmony.Pro.1.0.0  ( 53 MB )

80 MB

MIDI Printer: Generate Endless Melody Ideas in Seconds

Cymatics MIDI Printer composes an endless stream of original 8-bar MIDI compositions in the exact genre and key you’re working in, so you never stare at a blank piano roll again.

A Fresh Idea Every Time You Hit Print
MIDI Printer composes an endless stream of original 8-bar MIDI compositions in the exact genre and key you’re working in. Every print is a brand new composition, crafted on the spot from the songwriting DNA of Cymatics’ top melody creators.

With built-in originality protection, it guarantees you’re never handed a repeated composition.

Preview instantly with the built-in player, flip through your print history, save your favorites, nudge the octave, and drag the winner into your DAW in seconds.

How does MIDI Printer work exactly?
Never start with a blank piano roll again. MIDI Printer composes an endless stream of original 8-bar melodies, chords, and progressions in the exact genre and key you’re working in: one button, instant results, drag straight into your project.

Every print is a brand new composition built from the musical DNA of over a thousand professionally written Cymatics ideas, so it sounds written, not randomized.

Built-in originality protection guarantees it never hands you a repeated composition.

Preview instantly with the built-in player, flip through your print history, nudge the octave, and drag the winner into your DAW. From “I need an idea” to “I’m producing” in under ten seconds.

Visit Developer Site

Cymatics.MIDI.Printer.1.0.0  ( 81 MB )

50 MB

MIDI Shredder Plugin Instantly Convert Audio into MIDI Notes in Seconds

Shred Any Audio Into MIDI Notes with 1 Click and Drag
MIDI Shredder is a creative utility plugin that instantly converts any audio into playable MIDI, all through the look and feel of an office paper shredder.

Just drag an audio file directly into the top of the machine and watch it get “shredded” into MIDI notes in real time, transforming melodies, vocals, basslines, guitar riffs, and samples into MIDI compositions you can completely reshape.

Whether you’re extracting inspiration from loops, rebuilding melodies from old ideas, finding basslines or generating instant starting points for new songs, MIDI Shredder turns audio into usable MIDI in seconds with a workflow that’s surprisingly addictive.

How does MIDI Shredder work exactly?
Drop any audio file into the shredder slot and MIDI Shredder automatically analyzes the pitch information inside your sound.

As the file feeds through the machine, the audio visually transforms into shredded MIDI note strips that spill out into the collection bin below, ready to drag directly into your DAW’s piano roll.

Use it to recover chords from samples, turn vocals into synth leads, convert guitar ideas into MIDI instruments, or completely reinterpret existing audio into something brand new.

It’s fast, satisfying, and one of those tools that instantly sparks new ideas the second you start throwing sounds into it.

Visit Developer Site

Cymatics.MIDI.Shredder.1.1.0  ( 50 MB )

8.08 MB

MIDI Generator for Electronic Music Producers. Generate melodies, basslines, chord progressions, and arpeggios instantly for Techno, Trance, Psytrance, Drum & Bass, House, and other electronic music genres.

Features
🎼 MIDI Generation
– Generate Melodies, Basslines, Chords & Arpeggios
– 10 Electronic Music Genres
– 50+ Production-Ready Presets
– Harmonic Timeline with genre-aware chord progressions
– Melody, Bassline & Arpeggio follow the active chord of each bar
– Style Variants (Profiles A–H) for every preset
– Phrase-Aware Musical Generation
– Scale-Aware Generation
– Beat & Scale Degree Weighting
– Smarter Harmonic Resolution
– Dissonance Budget for more musical note choices
– Duration Profiles per genre (Techno, House, Trance, DnB, Ambient, Cinematic…)
– Phrase Extensions for more natural loop endings
– Improved Chord Generation with automatic validation and safe fallback
– Pitch Variety Enhancement
– Melodic Gravity for More Natural Phrases

🎛 Creative Controls
– Groove Templates (Straight, Swing, House, DnB & more)
– Melodic Contours (Arch, Ascend, Descend, Plateau & Valley)
– Tension Macro for Instant Energy Control
– Deterministic Seed System
– Recreate & Share Ideas with Seeds
– One-Click Random Inspiration

🎹 Piano Roll & History
– Editable MIDI Piano Roll
– Real-Time Pattern Editing
– Manual Edit Protection
– NEW: History Panel
– Save up to 20 favorite patterns
– Pattern labels
– Mini MIDI previews
– Play / Stop saved patterns
– Drag & Drop directly from History
– Edit saved patterns
– Delete saved patterns
– History automatically saved inside your DAW project

🔊 Audio Preview
– Built-in Audio Preview
– Genre & Role-Based Preview Instruments
– Dedicated sounds for Melody, Bassline, Chords & Arpeggio
– Improved stereo width and detuned synth sounds
– Voice-specific EQ
– 32-voice preview engine
– Reverb & Limiter
– Host-Synced Playback & Metronome

🚀 Workflow
– MIDI Drag & Drop Export
– Lightweight CPU Usage
– Windows & macOS Support
– Free Download

Visit Developer Site

ABYZOR.Genesis.1.7.0  ( 8 MB )

18.6 MB

Gliff is a MIDI effect for turning incoming notes into guitar-aware chord information and clean polyphonic MIDI. Put it on a MIDI track, play from a keyboard, MIDI guitar, DAW clip, or the on-screen fretboard, and it names the chord while showing a playable fretboard shape.

The important distinction is that Gliff is not an audio-to-MIDI converter. It does not listen to a recorded guitar signal; instead, it translates MIDI note data into guitar voicings, inversions, slash-chord interpretations, and output that can be dragged back into a DAW clip.

That makes it useful when you already know the harmony but want it to behave more like something a guitarist could play. Alternate tunings, custom tunings, capo modeling, real-time chord naming, and a one-bar MIDI drag workflow make it a compact helper for producers sketching guitar-like parts, guitarists using MIDI hardware, and keyboardists who need better fretboard logic inside a normal writing session.

Key Features
– Names held MIDI chords in real time, including slash chords, inversions, and partial matches.
– Shows incoming harmony as both a fretboard shape and a piano view for quick translation between instruments.
– Writes clean polyphonic MIDI output that can be routed to other instruments in the DAW.
– Lets users click the on-screen fretboard to audition voicings through the plugin’s MIDI output.
– Supports drag-and-drop of a one-bar MIDI clip from the plugin into the DAW timeline.
– Includes built-in standard, drop, DADGAD, open, ukulele, custom tuning, and capo options.
– Ships as macOS VST3/AU, Windows VST3, and a Max for Live device for Ableton Suite.

Visit Developer Site

gliff.1.5.0  ( 12 MB )

5.58 MB

Parallel Sequence Engine. This plugin mixes the best properties of an arpeggiator and a multitrack sequencer by allowing you to trigger up to five independent parallel sequences. Each sequence loops over one bar in the host’s tempo, but it starts when the assigned MIDI note starts and loops as long as it is held.

MIDI Note management is exactly as would be with an arpeggiator. First incoming note routes to the first sequence module, second note to the second, and so on. And just like an arpeggiator, notes that you release will force a re-assignment.

Example: Incoming notes C3, E3, G3
C3 plays in the first sequence, E3 in the second, G3 in the third. E3 is released. Now G3 plays in the second sequence.

Each sequence module’s steps can be freely set between 1 and 32. This allows for polyrhythms. Each step of each sequence has independent settings for MIDI channel, velocity and gate length parameters.

Everything can be tweaked at runtime, allowing experimentation.

Usage:
– Left click on a step to switch a step on or off.
– Right click on a step to switch parameter editing on or off.
– Left click and hold on a control to change value.

Visit Developer Site

Plexum.1.2.0.0  ( 5 MB )

14.7 MB

Audio-to-MIDI Transcription Plugin. You play or sing something — it listens, analyses the audio, and converts it into clean MIDI notes that you can use in your DAW.

Drop in a guitar recording, a piano melody, a vocal line, a bass riff, or even a full instrument — Semedo Notes figures out which notes were played, when they started, how long they lasted, and how loud they were. It turns all of that into a MIDI file you can drag straight into your project.

No keyboard skills needed. No MIDI controller needed. Just play or sing.

AUDIO TO MIDI — THE CORE
Drop any audio file onto the plugin — WAV, AIFF, FLAC, MP3 or OGG — and Semedo Notes analyses it and produces a full MIDI transcription in seconds. It works on single instruments: guitar, piano, bass, voice, strings, flute, synth lead — any source where individual notes can be identified.

LIVE RECORDING
Hit the REC button while your DAW is running. Semedo Notes captures your live input — microphone or instrument — for up to 2 minutes and transcribes it the moment you stop. No need to bounce to a file first.

GHOST NOTE FILTER (unique to Semedo Notes)
This is what sets Semedo Notes apart from similar tools. Every AI pitch detector generates false notes — overtones, room noise, bleed, and artefacts that appear in the MIDI as notes that were never actually played.

The Ghost Filter automatically removes these:
– It detects and removes octave overtones (harmonics that sound above the real note but were not played independently).
– It removes low-confidence notes that are too quiet relative to the real notes in the same moment.
– It enforces a polyphony limit so chords stay clean instead of turning into a wall of unwanted extra voices.

The result is MIDI you can actually use without spending an hour deleting wrong notes by hand.

PURITY CONTROL
A single knob controls how aggressively the plugin cleans up the result. Turn it down to keep everything the model detected. Turn it up to keep only the notes the plugin is most confident about. You stay in control of the balance between completeness and cleanliness.

PITCH BEND / EXPRESSION EXPORT
Semedo Notes detects the micro-pitch movement in your playing — vibrato, slides, bends — and can write that expression into the MIDI file.

Three modes:
– Off — flat MIDI notes, works with every instrument.
– Mono safe — adds pitch bend for single-note lines where bends cannot conflict with each other.
– Per-note MPE — each note gets its own MIDI channel and its own pitch wheel. Perfect for MPE-capable instruments like Serum, Pigments, ROLI instruments, and MPE-aware samplers. The MIDI file declares a proper

MPE lower zone so it plays back correctly without manual setup. Bend range: ±2, ±12 or ±24 semitones, written into the MIDI via RPN messages so the receiving instrument knows how to interpret the data.

TIME QUANTIZE
Align the transcribed notes to your project’s tempo grid automatically. Semedo Notes reads the BPM directly from your DAW — no manual entry needed while a project is open. Choose your grid (1/4, 1/8, 1/16, 1/32, triplets) and set how strongly the notes snap to it: 0 leaves timing completely raw, 1 locks everything to the grid. If no DAW tempo is available (standalone use or imported files), a manual BPM fallback is available.

SCALE FILTER
Constrain transcribed notes to a musical scale. Choose any root note and scale type: Major, Minor, Dorian, Mixolydian, Lydian, Phrygian, Locrian, Blues minor, Minor pentatonic, Major pentatonic, Melodic minor, Harmonic minor, Harmonic major — 14 scales in total. Notes that fall outside the chosen scale are removed, not moved. Semedo Notes never invents notes or shifts pitches to force a result.

NOTE RANGE FILTER
Set a minimum and maximum MIDI note to ignore anything outside the range you care about. Useful for filtering out low-frequency noise that gets transcribed as bass notes, or keeping only the range of a specific instrument.

MIDI DRAG AND DROP
When the transcription is ready, drag the MIDI clip directly from the plugin into a MIDI track in your DAW. No export dialog, no file browser — just grab and drop it where you need it.

MIDI EXPORT
Save the clean MIDI to a file at any point. The export button opens a standard save dialog. The exported file includes the correct tempo, note velocities, pitch bend data (if enabled), and RPN messages for MPE receivers.

MIDI OUTPUT
Semedo Notes sends its MIDI output to your DAW in real time. Route the plugin’s MIDI to any instrument track and preview the transcription through any sound you like.

BUILT-IN PREVIEW INSTRUMENT
A lightweight built-in synthesiser lets you hear the transcription inside the plugin without any additional routing. 32 voices, clean sine wave with smooth attack and release, soft limiter to prevent clipping. When MPE is active, pitch bends are played back correctly per voice in the preview too.

WAVEFORM DISPLAY
The original audio waveform is displayed above the piano roll so you can see exactly which notes correspond to which part of the recording. The waveform stays aligned with the notes and the playback head as you zoom and scroll.

PIANO ROLL
A full piano roll shows all transcribed notes with velocity shading — louder notes appear brighter. Zoom in and out with the mouse wheel. The playhead moves in real time during preview playback.

RE-ANALYSE FROM ORIGINAL
The original audio is kept in memory after transcription. If you change the Detection Confidence or other model parameters, Semedo Notes can re-run the full analysis from the original recording without you having to drop the file again or re-record.

LOW CPU USAGE
All heavy processing — audio loading, resampling, AI inference — runs on a background thread. The audio thread does no analysis work, no memory allocation, and holds no locks. Your DAW stays responsive and dropout-free while Semedo Notes works.

SAFE DURING PLAYBACK
Semedo Notes never interrupts your DAW’s audio output. Analysis runs in the background. Results appear in the piano roll when ready. You can keep playing and recording in your project the whole time.

Visit Developer Site

SemNotes.1.0.17  ( 14 MB )

5 MB

This is a tool that allows you to change the sound timing of drums you have created by playing them back and forth.

The pronunciation timing fluctuates according to the pitch and velocity settings of the MIDI signal,

Bass drums are pronounced earlier only when they become stronger.

– The hi-hat is pronounced slower only when it becomes weaker.

– The snare is always pronounced a little slower.

Grooves can be adjusted without manually fine-tuning MIDI tracks.

– Even if quantization is performed by changing swing values, etc.,

4D pocket groove designer can maintain displacement and randomization.

– Because it functions without moving the sound position on the sequencer,

the first note of a beat is pronounced a little before the beat. This makes it easy to edit drum tracks in which the first note of the beat is pronounced a little before the beat.

The goal is to make drums groove more naturally without manual MIDI editing.

Only Tested in Windows Studio One (May not work properly in other DAWs and Mac OS)

Visit Developer Site

4D_pocket_groove_designer  ( 5 MB )