Vallaren DTR-48 GET MUSE EQ
VALLAREN / PLUG-INS / Muse EQ
AI-Assisted Analog-Voiced EQ · 8 bands · 9 voicings

An equalizer that sounds like a record.

Muse combines nine hand-modeled analog voicings — Model 73, 81, 84, 80 series, Passive EQ, 4K-E, Q Eq, Range A — with the precision of a modern visual workflow, and hands a real AI mix engineer the first move. Press AI Assist and Gemma 4 writes a chain of voicings and band moves into the curve. Press De-ess and the same engine dials in a complete dynamic de-esser — frequency, Q, threshold, ratio, depth, attack, release — so harsh "S" and "SH" transients get solved on a single button.

GET MUSE EQ — $39
Version1.0
Voicings9 per-band
Bands8 · HPF · LPF
AILocal Gemma 4
De-essDynamic · AI-tuned
FormatsAU · VST3 · macOS
Dreamlab Muse EQ — plugin interface
01 / Nine voicings, per band

Most EQs give you one flavor. Muse gives you nine.

Each of the eight bands can use its own voicing. Every voicing reshapes the underlying biquad coefficients — these are not EQ presets, they are different filter shapes. Reach for the flavor that actually fits the move.

Classic

Model 73

The Class A, transformer-coupled 1970 voice with custom-wound input and output transformers — meaty low-end, silky high-frequency bloom, and rich harmonic saturation that adds body to vocals, drums, and electric instruments.

Classic

Model 81

The faster, tighter 1972 sibling of the same module family — discrete Class A/B, still transformer-coupled but noticeably more modern and controlled. A staple for drums, bass, and guitars.

Classic

Model 84

A Class A 1970s module with a different transformer winding — a subtly different harmonic distribution from the other modules in the family, plus a high-Q mid switch for surgical work with the same Class A timbre.

Classic

80 Series

The channel strip from the big-iron 1970s console — same Class A DNA as the module family, but with larger transformer character that gives it airier top end and the wider bus glue of the console voice.

Classic

Passive EQ

A passive two-band tube program EQ where harmonic character comes from the tube makeup amplifier, not a transformer — broad musical curves and distinctly different from everything else in the library. Implements the classic boost-and-cut trick: boost and cut at the same frequency and the band widens and shifts exactly the way the real hardware does.

Classic

4K-E

Deliberately not transformer-coupled — solid-state and VCA-based, with the punchy, aggressive, in-your-face midrange that defined eighties pop and rock records. Especially drums and master bus.

Classic

Q EQ

The defining Proportional-Q behavior: bandwidth narrows as gain increases. Small moves stay broad and musical at Q ≈ 0.3–0.5; large moves tighten into punch at Q ≈ 1.5–2.5 — the single most important sonic fingerprint of the original.

Classic

Range A

Inductor-based rather than capacitor-based — which is what produces its sweet, musical character and signature high-end "bloom." Not a simple boost but an extended, musical shimmer — the reason inductor EQs stayed in front of vocals right through the analogue era.

Digital

Clean

No transformer, no tube, no inductor, no saturation — bit-exact pass-through when gain is zero. For surgical cuts on muddy resonances, precise de-essing, or reference mixing where any coloration would be wrong.

02 / Features

Every layer tuned for musicality, not metrics.

F.01

A local AI mix engineer, on your machine

The AI Assist button is powered by Gemma 4 running locally through llama-server. Nothing leaves your computer, nothing costs anything per click, and the model responds in roughly the time it takes to re-listen to the clip. Gemma sees real spectral and loudness features, knows every voicing in the library and what each one is good for, and returns a complete chain move — global character, per-band frequency / gain / Q / type, optional per-band character overrides, plus HPF / LPF and transformer-drive suggestions. The same engine powers the dedicated De-ess button, which runs a focused sibilance analysis and writes a full dynamic de-esser recommendation into the chain.

Local · private · no per-click cost
F.02

A dynamic de-esser built into the signal chain

Not bolted on — a first-class DSP stage at the end of the chain. Architecturally the classic split-and-subtract design: a wide sidechain high-pass decides when to duck by watching for sibilant energy, while a narrow-Q bandpass on the reduction path decides what to duck by surgically carving only the harsh core. A one-pole envelope detector with independent attack and release controls the ducking amount — mathematically equivalent to a dynamic peak filter at the target frequency, but with fixed filter coefficients, so there is no zipper noise. The red-orange DE handle sits on the curve next to HPF/LPF, with a translucent notch that pulses downward whenever an "S" triggers ducking.

Split-and-subtract · live GR meter · freq 4–10 kHz
F.03

Nine hand-modeled voicings, selectable per band

Reach for the Model 73 or 84 for thick, Class A transformer warmth on vocals and bass. Drop the  4K- E on drums for that in-your-face pop / rock character. Put the Passive EQ on the low end for its signature boost-and-cut trick — boost and attenuate at the same frequency and the band widens and shifts exactly as the hardware does — or the Range A on guitars for the inductor bloom that gives layered rock guitars their air without harshness. Each voicing reshapes the underlying biquad coefficients. Each of the eight bands chooses its own flavor independently.

Classic · 70's Circuits · British EQ · Passive EQ · Modern · Clean
F.04

Analog-style two-stage gain staging

Before any internal stage sees the signal, the plugin attenuates by the peak gain of the current curve plus a three-decibel transient safety margin — accounting for the brief overshoot high-Q biquads produce on percussive material. After the chain, the plugin reapplies makeup gain in one of three musically distinct modes. Off is pure additive. Soft (default) weights the makeup by the RMS energy of the curve, loudness-matching broad shelves. Hard leaves the peak uncompensated, turning the curve's peak into the new zero-dBFS reference. A level-matched Bypass button tracks the current makeup gain live, so A/B compares tone — not "quiet dry" vs "loud EQ'd."

Off · Soft · Hard · level-matched Bypass · tanh soft-clip at −1 dBFS
F.05

The transformer that doesn't ruin transients

The input saturator is built on a threshold-linear shape that stays completely transparent below −4.4 dBFS, then softly rolls peaks above the knee using a tanh curve. Deliberately not a sigmoid — sigmoid saturators quietly compress transients by several decibels even at modest drive, which is exactly why most "analog-modeled" plugins sound dull on drums. A small positive-swing asymmetry adds second-harmonic warmth. At drive 0, the stage is a bit-exact pass-through.

Threshold-linear · 2nd-harmonic asymmetry
F.06

Pro-grade curve editor

The spectrum analyzer, frequency grid, band regions, summed curve, draggable nodes, HPF/LPF handles, and de-esser notch all paint in a single pass — so the curve never flickers or gets wiped by neighboring repaints. Every band has a draggable node with hover-linked highlighting in the band strip below, trackpad-friendly frequency and Q gestures, a click-through popover for precise entry, and color-coded translucent shaded regions that mirror the eight-band palette. The DE handle opens a dedicated de-esser popover with live GR metering; the popover and the band popover are mutually exclusive so the UI never stacks floating panels.

Drag nodes · scroll/pinch Q · ⌥-drag fine Q · live GR
F.07

Diagnostic honesty, session persistence

When something odd happens on Gemma's side, the plugin surfaces the raw reply in an in-plugin diagnostic dialog — you see exactly what the model said rather than a silent failure. Three layers of safety clamps guard the audio path, so even under prompt regression the model cannot push the plugin into unsafe territory. Every parameter — eight bands, HPF, LPF, input/output trim, saturator drive, auto-gain mode, bypass state, full de-esser configuration — round-trips cleanly through Logic and any other DAW session. The state format is version-tagged; old sessions missing fields added in newer builds fall back to safe defaults.

03 / Controls

Per-band precision, global character.

Per-band

Eight bands · each independent
  • FrequencyDrag node or type value
    20 Hz – 20 kHz
  • GainCut or boost
    ±18 dB
  • QScroll / pinch · ⌥-drag for fine
    0.10 – 10.0
  • TypeBell · shelf · notch
    5 shapes
  • VoicingPer-band character override
    9 models
  • Solo / BypassAudition or mute band
    per band

Global

Shared across the plugin
  • AI AssistGemma 4 · four-second listen
    one click
  • De-essAI-tuned dynamic de-esser
    one click
  • Global characterDefault voicing for the chain
    9 options
  • Input trimPre-chain attenuation
    ±24 dB
  • Auto-makeupGain-stage mode
    Off · Soft · Hard
  • BypassLevel-matched A/B
    honest A/B
  • Output trimPost-chain attenuation
    ±24 dB
  • HPF / LPFDraggable handles at curve edges
    bypassable
  • Transformer driveThreshold-linear saturator
    0 – 100 %
  • PresetsFactory library + AU program interface
    curated
  • ClearReset to musical defaults
    one click
Default band layout on Clear: 60, 150, 350, 800, 1600, 3500, 7000, 12000 Hz — bypassed HPF and LPF, zero drive, zero trim.
04 / Engineering

Musical by architecture.

Gain staging

Two-stage pre-attenuate / makeup

Pre-attenuate by peak curve gain plus 3 dB transient margin, then reapply makeup after the chain. No hidden internal clipping behind clean-looking input and output meters.

Safety

tanh soft-clip at −1 dBFS

A gentle final clipper guarantees nothing hotter than full scale ever leaves the plugin — even under wild settings or prompt-regressed AI moves.

Saturator

Threshold-linear, not sigmoid

Below −4.4 dBFS the input stage is transparent. Above the knee, a tanh curve colors peaks with 2nd-harmonic warmth. Transients survive intact — which is why it actually sounds like hardware on drums.

De-esser

Split-and-subtract architecture

Wide sidechain high-pass detects sibilance; narrow-Q bandpass on the reduction path surgically ducks only the harsh core. One-pole envelope detector with independent attack and release. Mathematically equivalent to a dynamic peak filter — but with fixed coefficients, so no zipper noise.

AI Safety

Three clamp layers

Every move Gemma returns is passed through three layers of range / sanity / stability clamps before touching audio. The model cannot push the plugin into unsafe territory.

Persistence

Version-tagged XML state

Every parameter — including the full de-esser configuration and armed-vs-bypassed flag — serializes on save. Old sessions missing fields added in newer builds fall back to safe defaults; future fields in newer sessions are tolerated by current builds.

Local inference

Gemma 4 via llama-server

Runs on your machine. Nothing leaves the computer, nothing costs per click. The model sees real spectral and loudness features — not opaque embeddings — and responds in seconds.

Rendering

Single-pass curve editor

Analyzer, frequency grid, band regions, summed curve, nodes — all painted in one pass. No flicker, no paint-ordering bugs between sibling components. The curve is always coherent with what you hear.

05 / At a glance

The short version.

8 bands
Parametric, fully independent
9
Hand-modeled voicings
±18 dB
Per-band gain range
4 sec
AI Assist listen window
Gemma 4
Local inference · llama-server
3 modes
Auto-makeup · Off / Soft / Hard
De-ess
AI-tuned · live GR meter
−4.4 dBFS
Saturator knee
−1 dBFS
Final soft-clip ceiling
Audio Unit (AU)
VST3
macOS · native
Local AI · no data leaves
Automation-ready
Why it matters

Most EQ plugins give you either transparency or character. Muse gives you both — with an AI that understands the difference, and a dynamic de-esser that fixes the one problem a static EQ fundamentally cannot solve.

"Musical" is an architectural commitment — from the per-band voicing library, to the gain staging, to the saturator curve, to the passive EQ’s cross-band interaction, to the way Gemma picks a different flavor for a low shelf on a kick than for a surgical cut on a muddy vocal. Every decision was made in service of one question: does this sound like a record?

07 / More from DreamLab

The rest of the rack.

Dynamics

Multiband

Phase-coherent 4-band mastering compressor with interactive spectrum and A/B peak-hold.

Modulation

DuoPhase

Dual 6-stage optical phaser with two sweep generators. Deep series swirls or wide parallel stereo motion.

Synthesizer

Vorion

Three oscillators, eight voices, and a modulation matrix built for trouble.

Overdrive

ScreamDriver

Circuit-modelled three-stage overdrive with tube-style compression, tape saturation, BBD chorus, pitch-bending echo, spring tank and a zero-latency cabinet loader.

Granular atmosphere

Lush

A granular atmosphere engine — Drone, Reverse, Shimmer and Gliss modes, pitch-tracked grains, and a Freeze that holds a cloud indefinitely.

Rhythm composer

DreamLab Drum Machine

Sixteen voices on a step sequencer that works the way the originals did — per-voice tune, decay and filter, accent per step, and producer-style swing.

Neural amplifier

Zelos

Neural amp capture and playback. Load a model or capture your own rig. Free · public beta.

Lo-fi

DreamLab Lo-Fi

Seven lo-fi modules on one panel — saturation, wow and flutter, bit crush, band limiting, dropouts, ambience and a ducked noise bed, under one Amount macro.

Delay

DreamLab Delay

Stereo analog echo and ducking delay — tape and lo-fi character, repeat filtering, ping-pong, and program-dependent ducking.

Currently viewing

Muse EQ

AI-assisted equalizer · 9 analog voicings per band · Gemma 4 local inference.

Synthesizer

DDM S1

64-voice hybrid subtractive + 6-op FM polysynth with 8 algorithms.

Reverb

Convolution Reverb

Multi-engine reverb: a true-stereo convolver plus six algorithms, with shimmer, freeze, duck, and delay-into-reverb.

Saturation

SAT 456

Physically-modeled tape saturation — harmonic warmth, transient punch, smooth top end, and wow/flutter motion from a single insert.

Synthesizer

Synth CH388

Six-voice virtual analogue polysynth — rock-stable oscillator, creamy four-pole filter, and a two-mode stereo chorus. Zero latency, one panel.

Spatializer

Binaural Panner

3D placement panner with crosstalk cancellation. Works on headphones and real stereo monitors.

Restoration

Adaptive Noise Reducer

Real-time spectral suppression + envelope gate. Dual-stage attenuation with live activity meters.

Networking · free

AudioOverNetwork

AES67 / ST 2110 audio-over-IP between DAWs. Sub-5 ms on gigabit. MIT-licensed, open source.

Spectral filter

Sound Morph

Morphing spectral filter · 5 anchor shapes · dual-LFO motion · studio shimmer.

Modulation

D Chorus

Stereo BBD chorus and pitch shifter. Wide, warm, and wild. AU · VST3.

Metering

LUFS Master

BS.1770-4 loudness metering, phase scope, and a true-peak brickwall limiter. AU · VST3 · Standalone.

Pitch

Pitch Corrector

Real-time pitch correction — natural intonation to hard-tune, with formant preservation. AU · VST3 · Standalone.

Drum trigger

Drum Trigger

Audio transient to MIDI generator. Tighten drums without losing groove. AU · Standalone.

Companion tool

Impulse Lab

Build your own impulse responses — design a space or capture a real one, deconvolve, and load it straight into Verb.

MIDI generator · free

Vibe Generator

A MIDI plug-in that writes full ambient, neo-soul and lo-fi songs across five parts. Free · public beta.

FORMATS
AU · VST3
ARCHITECTURE
Universal — Apple Silicon and IntelAPPLE M-SERIES RECOMMENDED
SECURITY
Signed and notarised by Apple
REQUIRES
macOS 10.13

Muse EQ is an original Dreamlab Digital Media product. It is not affiliated with, endorsed by, or derived from any hardware or software manufacturer. Full licence terms at dreamlabnet.com/daw2/eula.html.