74.6

Audit Your Audio with Confidence. A powerful, open-source toolkit for audio analysis and playback. Verify lossless quality, detect AI-generated tracks, and explore your library with a built-in hi-res player and advanced EQ.

Fake Lossless Detection
Identifies files that claim high quality but are actually upsampled from lower bitrate sources via FFT spectral analysis.

Spectral Analysis
4096-point Hanning-windowed FFT determines the true effective frequency ceiling of your audio with precision.

Clipping Detection
Digital clipping scan with percentage and sample-count reporting. Thorough mode detects clipping even when scaled down.

MQA Detection
Identifies MQA and MQA Studio encoded files, reports original sample rate and encoder information.

AI Audio Detection
Scans metadata, byte patterns, C2PA provenance, AI watermarks, and audio signals to detect AI-generated music. The unified AI column shows Yes / Possible / No verdicts with confidence scores, combining standard detection, SH Labs API, and experimental spectral analysis.

Rich Export
Export to Excel, CSV, Text, PDF, or Word — all matching your DataGrid layout with styled formatting.

Fake Stereo Detection
Identifies mono-duplicate and near-mono files using inter-channel correlation analysis.

Waveform & Spectrogram Compare
Compare audio files in time-domain waveform and full-resolution spectrogram. Features stacked, overlay (with pixel-level difference heatmap), and wipe modes. Ideal for verifying remasters and lossless authenticity.

Batch Tools
Batch rename with auto-rename from tags, duplicate detection, metadata stripping, playlist import, and cue sheet parsing for virtual tracks.

Search and Scrobbling
Connect with music services to discover tracks, search across platforms, and automatically scrobble your listening history.

AcoustID Fingerprinting
Identify unknown tracks by audio fingerprint via the AcoustID and MusicBrainz database. Auto-downloads fpcalc — no setup required.

Full Metadata Editor
Edit, add, or remove tags across all supported formats — ID3, Vorbis, APE, M4A. Includes a Batch Metadata GUI to enrich missing tags from MusicBrainz.

Loudness & Dynamics
Integrated LUFS/LKFS per ITU-R BS.1770, True Peak (dBTP) at 4× oversampling, Dynamic Range (DR), and Replay Gain.

Synced Lyrics & Karaoke
Multi-provider synced lyrics with word-by-word karaoke highlighting, real-time translation, and click-to-seek. Sources include LrcLib, Netease, Musixmatch, embedded tags, and local .lrc files.

Now Playing Panel
Immersive full-screen Now Playing experience with album-cover color-matched glows, customizable layout (cover, title, artist, visualizer positions), and gradient backgrounds extracted from art.

Metadata
– Artist & Title
– Sample Rate & Bit Depth
– Channels & Duration
– File Size & Bitrate
– BPM Detection
– Replay Gain
– True Peak & LUFS
– Dynamic Range

Spectral
– 4096-point FFT
– Hanning Window
– 100 Analysis Segments
– Effective Frequency Cutoff
– Band-Energy-Drop Method
– Codec-Aware Estimation

Detection
– Fake Lossless / Upconvert
– Digital Clipping
– Scaled Clipping
– MQA / MQA Studio
– AI-Generated Audio
– Fake Stereo Detection
– Silence Gap Detection
– Optimizer Processing

AI Detection
– Metadata Tag Scanning
– Raw Byte & UTF-16 Patterns
– C2PA / Content Credentials
– AI Watermark Identifiers
– SH Labs API Integration
– Experimental Spectral (7 Checks)

Built-in Player

A full audio player, built right in
No need to switch apps. AudioAuditor includes a complete audio player with features rivaling standalone players — including a full-screen Now Playing experience with synced lyrics.

✅ Hi-Res Audio
Native 96/192 kHz playback with automatic fallback resampling

✅ 10-Band Parametric EQ
32 Hz to 16 kHz with ±12 dB per band, soft clipping protection, built-in presets, and custom profile saving

✅ Spatial Audio
HRTF crossfeed, interaural time delay, head shadow, and early reflections

✅ Crossfade & Normalization
Configurable crossfade (1–15s) with 4 curves (Linear, Equal Power, Natural, Sequential), gapless playback, and peak-based audio normalization

✅ Shuffle, Queue & Loop
Fair Fisher-Yates shuffle, drag-and-drop queue management, and session-persistent loop modes

✅ 6 Visualizer Modes
Bars, Mirror, Particles, Circles, Scope & VU Meter at 60 FPS with independent theming

✅ System Integration
Windows media overlay support (SMTC) — now-playing info in volume OSD, media key controls, and third-party overlays

✅ Now Playing Panel
Immersive full-screen Now Playing with ColorMatch Eyedropper, 4 animated backgrounds (Snow, Rain, Leaves, Underwater), and customizable layouts

✅ Floating Mini Player
Compact player with playback controls and an independent visualizer

✅ Your Wrapped Dashboard
A rich, range-aware statistics dashboard with interactive charts and local listening tracking

✅ Synced Lyrics & Karaoke
Multi-provider lyrics (LrcLib, Netease, Musixmatch, embedded tags, local .lrc) with word-by-word karaoke highlighting and real-time translation

Lossless
FLAC, WAV, AIFF, ALAC, APE, WV, TTA, TAK

Lossy
MP3, AAC, M4A / M4B / M4R, OGG, OPUS, WMA, MP2, MPC, AC3, AMR, SPX

Container & Specialty
DSF / DFF (DSD), MP4, MKA, WEBM, 3GP / 3G2, AU / SND

Discord Rich Presence
Shows currently playing track, artist, elapsed time, and duration progress bar. Fetches album art from Last.fm with play/pause state icons and auto-reconnection.

Multi-Service Scrobbling
Full API integration with Last.fm, Libre.fm, ListenBrainz, and Maloja. Configurable thresholds, blacklisting, and simultaneous scrobbling.
Music streaming service integrations including Spotify, Tidal, Qobuz, Apple Music, YouTube Music

6 Music Service Slots
Choose from: Spotify, YouTube Music, Tidal, Qobuz, Amazon Music, Apple Music, Deezer, SoundCloud, Bandcamp, Last.fm, or any custom URL!

🌐 Optional Network Calls
No internet activity except when you explicitly use music search, Discord RPC, Scrobbling services, or the optional SH Labs AI detection. Offline Mode disables everything with one toggle.

< Minimal Disk Usage
Only small settings files and temporary archive extractions are written to disk. Analysis data lives in memory and temp files are cleaned up automatically.

📄 Local Logging Only
All optional logs are stored safely on your machine. No remote crash reports. No usage tracking.

🤖 Zero AI Training
Nothing analyzed or played is ever used to train generative AI. Period.

Visit Developer Site

AA.1.8.0  ( 75 MB )

17.3 MB

Interactive Graphic Score & Serial Music Generator. POINTILLISME is a browser-based generative quartet built entirely with the Web Audio API — a single self-contained HTML file, no installation, no dependencies.

Four independent oscillators function as four voices in a serial ensemble. Each voice runs seven independent series that advance at their own pace: pitch (cycling through all 48 row forms — Prime, Inversion, Retrograde, Retrograde-Inversion, each at 12 transpositions), register, duration, dynamics, envelope (ADSR), spatial position (stereo pan), and ornament. Because the series lengths are coprime, the global combination does not repeat for millions of events — an arithmetic property that produces the absence of repetition and hierarchical landmarks characteristic of integral serialism, not as an imposed rule but as an emergent consequence of the system itself.

The ornament series draws from ten types rooted in traditional and extended technique: trill, mordent, gruppetto, appoggiatura, acciaccatura, glissando, tremolo, arpeggio, cluster, or none. Each ornament is rendered both in sound and as a graphic glyph on the scrolling score.

The visual output is a graphic score in the tradition of Stockhausen’s Studie II and Earle Brown’s coordinate notation: time flows horizontally, frequency vertically, each event is drawn as an envelope-shaped figure whose width is its duration, height its dynamics, and profile its ADSR contour. The four voices are distinguished by colour (sanguine, ultramarine, green earth, ochre) and by texture (smooth fill for sine, horizontal hatching for triangle, serrated edge for sawtooth, vertical striations for square). Ornament glyphs appear above or beside each event using conventional notation symbols.

Compose with it
Press play and POINTILLISME generates a continuous, non-repeating serial texture. Each voice has its own waveform selector (sine, triangle, sawtooth, square) and can be independently re-seeded or muted (tacet). The base time unit controls the pace of the generative process.

When you hear material you want to keep, press record. The instrument captures the full stereo output and all note events simultaneously. When you stop recording, two export buttons appear:
– Save audio — WAV or AIFF, 16-bit or 24-bit, at your system’s native sample rate (typically 48 kHz). Ready to drop into any DAW.
– Save MIDI — a Standard MIDI File (Type 1) with four separate tracks, one per voice (OSC I through OSC IV), each on its own MIDI channel. Drag this into Ableton Live, Logic, Reaper, or any other DAW and each voice arrives on its own track. From there, you can assign your own instruments, edit individual notes, transpose voices, or use the serial material as a compositional skeleton for an entirely new piece. All ornament events (trill alternations, glissando steps, arpeggio tones, cluster notes) are written as discrete MIDI events, so nothing is lost in translation.

Perform with it
The freeze function (button or F key) stops time: the score halts, every event visible at that moment stays exactly where it is, and the canvas becomes interactive. Click any event and it plays — with its original waveform, envelope, ornament, and pan position. This turns POINTILLISME into a live instrument: you freeze a moment, explore its sonic content by clicking individual events in any order you choose, then unfreeze to let the generation continue. In a live performance context, alternating between generative flow and frozen, manually-triggered gestures creates a natural dialogue between algorithmic process and human decision.

Technical details
– Single self-contained HTML file — works in any modern browser, no server required
– Pure vanilla javascript and Web Audio API — zero external dependencies
– WAV encoder: PCM, little-endian, 16/24-bit
– AIFF encoder: PCM, big-endian, 80-bit extended sample rate field, 16/24-bit
– MIDI encoder: Standard MIDI File Type 1, 480 ticks/beat, conductor track + 4 voice tracks
– Stereo output with per-voice panning (serialized azimuth parameter)

Visit Developer Site

POINTILLISME  ( 17 MB )

30/229 MB

Play everything.
Capture anything.
Export anywhere.
NAM 9000 is two apps in one: a complete amp-sim rig — pedals, dual amps, studio EQs, cabs and FX — and a built-in capture studio that turns your real amps into standard .nam files. Yours to keep: load them in any NAM player, or export straight to NAM hardware. Both apps free.

A full rig, and the machine that builds rigs

Everything between your guitar and the mix
Paged like a real rig: pedalboard, amp head, rack, cab. No menus buried in menus.

The pedalboard
Gate, Tube Screamer, compressor, pitch, chorus, phaser, flanger, tremolo — plus the CAPTURE pedal: a second NAM engine that runs captured drive and fuzz pedals in front of your amp, simultaneously.

Two NAM engines at once — pedal capture into amp capture
Pedal captures from TONE3000 auto-route here
Tape delay and four-engine stereo reverb — room, hall, spring, plate — with trails bypass

The ampGlowing tube head, multi-capture cycling on the faceplate, Full/Lite quality toggle.
Stacked graphic and parametric EQ panels with live analyzers
Two studio EQs
10-band graphic rack and an 8-band parametric with HP/LP — stacked panels, both in the chain at once, each with its own live analyzer.
4×12 cabinet page
The cab
Classic-style 4×12 with IR loading. Amp+cab captures auto-bypass it — no double-cab mud.
Profile browser with brand folder tree
The library
Hundreds of profiles in a resizable brand-tree browser — search, favorites, folders, one-click load.

Your amp, immortalized
Profiling isn’t an external tool or a paid add-on — it’s a page in the plugin. Setup → Test → Capture → Train → Compare, all guided.

The capture studio
NAM 9000 plays the official NAM capture sweep through your rig, records what comes back, and trains a neural model on your own machine — GPU-accelerated, nothing uploaded anywhere. No NVIDIA GPU? One click exports the recording to TONE3000 and trains it free in the cloud. This is the actual capture scene from the plugin, cloning a rig on loop:

Capture amps, drive pedals, or full amp+cab rigs
Built-in level test and sweep calibration
A/B compare the clone against the real amp when it’s done
Output is a standard open .nam file — not a locked format
Export any capture for NAM hardware — Dimehead, HoTone, Darkglass and friends
Share on TONE3000 or keep your secret sauce private
YOUR RIG
THE CLONE
READY
waiting to record
🎸
Hook up your amp
Interface out to the amp, mic or load box back in
›
📡
Play the sweep
The plugin sends the calibrated capture signal through your rig
›
🧠
Train locally
The neural trainer runs on your own machine
›
⚡
your-amp.nam
Lands in your library, ready to play or share

Powered by community

250,000+ captures, zero cost
NAM 9000 speaks directly to TONE3000, the largest open library of neural amp captures. Search by gear, audition, one-click download — amps, pedals and full rigs route to the right slot automatically.

Search with gear and architecture filters (A1 / A2)
Pedal captures auto-route to the CAPTURE slot
Amp+cab captures auto-bypass the internal cab
Multi-capture tones download on demand as you cycle

Free. No account. No strings.
One email, both installers — Windows and macOS links land in your inbox. Free, no account, unsubscribe anytime.

⚙️ Requirements
Windows 10 / 11, 64-bit · ASIO interface recommended
macOS 11 (Big Sur) or later, Apple Silicon or Intel
VST3 / AU host (any major DAW) or run standalone

Visit Developer Site

NAM.9000.1.8.0  ( 327 MB )

57/91MB

Calm EverClean V2 is a free forever live-audio processor from Everworks, designed for feedback control, channel cleanup and fast real-world live workflow use.

This is not a trial, not a demo and not a subscription.
No watermark. No paid license. Free forever.

The Everclean Engine and the Calm Mamba neural network improves the performance and usability of the DeFeedback-Debleed-Denoise concept with a cleaner, more stable and more practical workflow for engineers working under real show conditions

It is built for live sound engineers, monitor engineers, system techs, venues, festivals, rehearsals and production environments where speed, stability and clean control matter more than unnecessary complexity.

* Feedback control and live channel cleanup
* Practical real-time workflow for live engineers
* Improved performance around the AlphaLabs DeFeedback concept
* Designed for real venues, rehearsals and festival use
* Simple and fast to operate
* Official freeware release
* No protection
* No demo limits
* macOS and Windows builds
* AU, VST3 and Standalone where available

Visit Developer Site

CALM.EverClean.V2.WIN  ( 54 MB )

CALM.EverClean.V2.MAC.Intel  ( 89 MB )

43 MB

Amp simulator with the latest neural modeling technology NAM A2
DustyTube 2 is a guitar and bass amp simulator plugin built on Neural Amp Modeler (NAM), an open-source project that uses deep learning to create highly accurate models of amplifiers and pedals.

Designed with simplicity and inspiration in mind, Dusty Tube 2 combines the power of NAM with a carefully curated set of features. All controls are accessible on a single page, creating a nostalgic, familiar environment that’s easy to use and inspiring to play. It’s free to download. I hope you find Dusty Tube a valuable addition to your toolkit and enjoy using it.

WHAT’S NEW
Upgraded to NAM A2 the latest neural engine with slim capability on the spot for loaded pedals and amps.
Built on a new framework. Better optimization and improved cpu usage.
Stereo input processing
Second pedal slot with mixing ability.
Mod section with essential FX including compressor, tremolo, vibrato, chorus, delay and room/spring reverbs.
preset manager to save your tones.
depth and low cut filters for amp.
Input/output calibrations for pedals and amp.
Unloading profiles.
Adoptive limiter in settings section
Neural saturation to boost captures harmonics
Sag block envelope follower to produce power amp compression, pick bloom and tube-like feel.
Additional Modern theme to classic DustyTube vintage style.
Supports:
Windows VST3/Standalone

Visit Developer Site

DustyTube2  ( 43 MB )

12 MB

Gateway is the plugin behind the NAM revolution. Load and play snapshot models of your favorite gear.

For everything you need to know about Neural Amp Modeler (NAM) Architecture 2 (A2): how it sounds, how it works, and how to use it, go to:
tone3000.com/guides/nam-a2-the-complete-guide

Visit Developer Site

Gateway-v1.0.1-win  ( 4 MB )

Gateway-v1.0.1-mac  ( 9 MB )

10.9 MB

Microtonal Piano Roll. Xenroll is a free and open-source piano roll plugin that allows you to define pitch with an accuracy of one cent.

– Define pitch with 1-cent accuracy (1200 EDO)
– Bend notes individually with visible control
– Generate pitch curves by singing or humming in real time
– Unique pitch memory model showing tonality, stability, and consonance
– Two dissonance submodels with visualization (compactness and beating)
– Clock diagram for notes, intervals, and chords in real time
– Import/export tracks, multiple instances with ghost notes
– MPE and MTS-ESP support for microtuning

Video examples

Visit Developer Site

XenRoll.0.4.3  ( 11 MB )

4.41 MB

Texture Loom is a free dual-sample atmosphere engine by S1gns Of L1fe, designed for creating evolving pads, drones, cinematic beds, textural layers, and ambient movement from your own audio.

Load two samples, blend between them, shape them with filtering and motion, then send the result into an onboard space/reverb section. Texture Loom is simple enough to start making sound immediately, but flexible enough for slow-burning sound design, generative ambience, and deep sample-based texture work.

Originally built as a Max for Live device, Texture Loom now also includes a native plugin version so it can live outside Ableton while keeping the same visual identity and workflow.

What It Does
– Dual-sample playback engine
– Load your own audio into Sample A and Sample B
– Start/Stop, Reverse, Start position, Level, Pan, and Rate controls
– RATE / SEMI mode for continuous speed control or semitone-based pitch movement
– Equal-power A/B blend control
– Blend LFO for slow crossfading motion
– Multimode filter with Lowpass, Highpass, Bandpass, Notch, Peak, and Allpass
– Filter cutoff, resonance, drive, and LFO modulation depth controls
– Two Motion LFOs with multiple shapes: Sine, Triangle, Ramp Up, Ramp Down, Random, Drift, and Glider
– Motion destinations for Blend, Filter, Space, Sample A Start, and Sample B Speed
– Onboard stereo space/reverb section with Size, Decay, Mix, and Width
– Master output with metering
– INIT button for returning to a clean blank state

Good For
– Ambient pads
– Evolving drones
– Cinematic transitions
– Experimental sample mangling
– Meditation and yoga music beds
– Underscore textures
– Sound design layers
– Turning field recordings, vocals, synths, guitar, noise, or found sound into atmosphere

Important
Texture Loom is an instrument, not an audio effect. Load it on an instrument track or software instrument channel. A blank instance makes no sound until samples are loaded. If you save a project with samples loaded, Texture Loom will restore those sample paths when the project opens again.

System Requirements
Max for Live version:
– Ableton Live 12
– Max 9

Native macOS version:
– macOS
– AU or VST3 compatible host

Windows version:
– Any VST3 compatible host

Visit Developer Site

Texture.Loom.1.0.1  ( 4 MB )

Size: 700mb

Turn any song into a drum kit

SampleMonsta is a FREE intelligent drum sampler that extracts, classifies and organizes drums from any song. Turn any song into a drum library in seconds.

Key Features:
Infinite Drum Source: Use any song as a sample library — literally any song becomes a playable drum kit,
Intelligent Classification: SampleMonsta identifies and groups similar drums automatically,
Color-Coded Zones: Visual organization makes finding the right sound fast and intuitive,
Export to DAW: Drag and drop your selected drums directly into your production workflow,
Constantly Evolving: New features and improvements added regularly,

How It Works:
Drag any audio file into SampleMonsta. The plugin analyzes the track, separates the drums using advanced separation technology, and organizes them into a playable, color-coded library. No more digging through endless sample folders or buying expensive sample packs.

Technical Specs:
Formats: VST3, AU,
Windows 10+ (VST3),
macOS Intel & Apple Silicon (VST3/AU)
Version: 1.5.0 (Windows & Apple Silicon) / 1.4.1 (Intel),

Visit Developer Site

Sample.Monsta.1.5.0.WIN  ( 692 MB )

Sample.Monsta.1.5.0.MAC  ( 659 MB )

Sample.Monsta.1.4.1.MAC.Intel  ( 510 MB )

53.3 MB

AudioDice: a creative VST host that randomizes plugin settings. You can load AudioDice in your favourite DAW as a plugin (VST/AU) and then load the effect or instrument of choice you want to randomise. Each parameter’s range and randomisation width can be controlled and you can randomize indiviidual parameter or all of them, even thousands. Control each parameter’s range and width (±100%) for spontaneous inspiration.

Visit Developer Site

AudioDice.1.1.1  ( 53 MB )