STEMDESKSTEMDESK
StemDesk separates your track into stems, builds a complete mix through real VST3 processing, and hands you the session — plus 17 professional plugins that work in Pro Tools and every major VST3 DAW.
Free tier includes the Compressor, Gate, and Limiter plugins — no account required.
Mix Targets — a real 24-track session: roles, LUFS, gain staging, rides & sidechain, local AI ready.
A complete desktop mixing pipeline that runs on your machine — your audio never leaves your computer.
Drop in a stereo 2-track and StemDesk splits it into 4 or 6 stems (vocals, drums, bass, guitar, piano, instruments) with Demucs-class source separation running locally on CPU — or start from your own stems folder, a Pro Tools session, or a reference track.
Every stem is measured — role detection, LUFS, spectral balance, key and BPM — against one of 150+ genre profiles, from trap and drill to gospel, rock, and Latin.
StemDesk builds the mix: per-stem processing chains, gain staging, pan, bus processing, sidechain ducking, vocal rides, and mastering — rendered offline through the same real VST3 plugins you use in your DAW. Every decision is deterministic and inspectable, with an optional local AI assistant for critique and adjustments.
Export stems, mix, and master as WAV, MP3, FLAC, or AAC — or take the whole session into your DAW: REAPER, Ableton Live, DAWproject, or a Pro Tools session built for you track-by-track.
New Mix Session — point at your stems or a 2-track, pick from 150+ genre profiles, go.
A full AI panel lives beside your session: critique, coaching, analysis, and generation — powered by a local model that runs on your machine. Your audio never touches the cloud, and the AI never silently edits your mix: every suggestion is shown, explained, and applied only when you say so. You can even start a brand-new session from a generated idea — "Create with AI" lives right on the setup screen.
Mix Critique — a graded report card: what's working, what's masking, and one-click "Apply Fixes".
Mix Coach — ask "why is my vocal buried?" and get session-specific moves, in dB, ready to apply.
AI Generate — smart suggestions in your song's key & BPM, local music generation, and a lyric writer.
Detects over-compression, clipping, noise, and spectral problems on every stem — each finding comes with a suggested fix.
Per-stem energy across seven bands with conflict highlighting, so you can see masking before you hear it. Plus arrangement density and bounce analysis.
Transcribes vocals in 11 languages (Whisper-class, local), detects phrases and breaths, and generates vocal ride automation from the phrasing.
Load a commercial reference next to your mix, compare them band by band, and optionally apply the match on render.
Generates preset chains for EQ, compression, saturation, multiband, de-essing, gate, and limiter from your session's measurements — and can even write preset files for FabFilter's Pro series if you own them.
Detects key, BPM, chord progression, and beat grid from the session — and generation uses them, so new ideas land in your song's key.
A real-time loudness meter with streaming targets built in — master to Spotify, Apple Music, or Amazon levels instead of guessing.
Save any mix's roles and override knobs as a .stemdeskprofile
file and re-apply it to a new session — your sound, carried between songs.
Built for Windows x64 in VST3 and AAX, validated at pluginval's maximum strictness, PACE-signed for Pro Tools. Three are free forever. Every interface below is a live capture of the real plugin.
The whole console channel in one window — 13 modules including EQ, compressor, gate, de-esser, saturation, delay and reverb, with a re-orderable chain and one-click AI configuration.
Eight character modes from Clean and Opto to Vocal, Drums, Bus and Mastering — with lookahead, program-dependent release, auto-makeup, parallel mix, oversampling, and a filtered external sidechain.
Wideband or split-band sibilance control with stereo / mid / side processing, external sidechain, and an audition mode that lets you hear exactly what's being removed.
Broadband noise reduction with threshold, depth, and attack/release envelope control, plus a mix knob for transparent parallel cleanup.
Tempo-synced echoes — straight, dotted, and triplet subdivisions — with ping-pong, diffusion, stereo offset, filters, and built-in ducking that keeps repeats out of the vocal's way.
Tune kicks, 808s, and toms to the song — automatic or manual note targeting, chromatic or scale-locked, with cent-level control and transient preservation.
8-band parametric with dynamic EQ per band, zero-latency or linear-phase processing, per-band stereo/L/R/mid/side routing, auto-gain, spectrum and sonogram analyzers, and a piano-key frequency strip.
Thickness on demand — chorus, micro-shift, Haas, and exciter engines with drive, tone, width and spread, for vocals and 808s that fill the speakers.
Clean gating with lookahead, hysteresis, hold and range control, a filtered external sidechain, and parallel mix — from drum tightening to hard chops.
Mastering-grade peak control: five styles from Transparent to Aggressive, oversampling, auto-release, LUFS loudness targeting, mid-side mode, and dithering.
Four bands with adjustable crossovers — linear-phase capable — each with the full compressor mode set, solo/mute per band, oversampling, and external sidechain.
Room, Hall, Plate, Chamber and Ambient algorithms with frequency-dependent decay, damping, tempo-synced pre-delay, freeze, and width control.
Eight saturation models — Tape, Tube, Diode, Transformer, Amp, Fuzz, Crush, Clean — with drive, feedback, tone shaping, filters, and oversampling.
Three-band width control with adjustable crossovers and a mono-bass switch — wide tops, focused mids, club-safe low end.
Independent attack and sustain shaping with speed control per stage and parallel mix — punch up drums or pull a room back.
Three repair stages in one pass — de-breath, de-click, and de-plosive — each with its own sensitivity and bypass. Raw vocal in, clean take out.
Real-time, formant-preserving pitch correction — key and scale lock, retune speed from hard-tune to natural glide, and an amount control for how deep it reaches.
The plugins work in every DAW that loads VST3 or AAX — REAPER, Cubase, Ableton Live, Studio One, Bitwig, FL Studio, and Pro Tools among them. Pro Tools is empirically tested end-to-end; VST3 hosts follow the VST3 standard, and every plugin passes pluginval strictness-10 host-lifecycle validation.
| Host | Plugin format | Status |
|---|---|---|
| Pro Tools 2022–2024 | AAX (iLok required) | Tested |
| REAPER 6/7 | VST3 | Standards-validated |
| Cubase 12/13 | VST3 | Standards-validated |
| Ableton Live 11/12 | VST3 | Standards-validated |
| Studio One 6 | VST3 | Standards-validated |
| Bitwig Studio 5 | VST3 | Standards-validated |
| FL Studio 21 | VST3 | Standards-validated |
| Logic Pro (macOS) | AU | v1.x roadmap |
Beyond hosting the plugins, StemDesk can hand your whole Auto-Mix to a DAW as a session:
| Target | What you get |
|---|---|
| REAPER (.rpp) | Full session — tracks, levels, pans, bus structure, and processing chains |
| Pro Tools (live) | StemDesk drives a running Pro Tools and builds the session track-by-track |
| Ableton Live (.als) | Session with stems, levels, pans, and song tempo |
| DAWproject | Open interchange format for Studio One and other supporting hosts |
One-time license, activated offline. No subscription, no iLok cost for VST3. Start free, upgrade when you need the full suite.
No. Separation, analysis, mixing, and the optional AI assistant all run locally. Nothing is uploaded, ever.
Only for the download and the optional one-time AI model fetch. Mixing, separation, and the plugins work fully offline.
The mix itself is built by deterministic, measurable DSP. The local AI suite sits on top — graded critique, a session-aware coach, per-stem diagnostics, vocal transcription, reference matching, preset generation, and music & lyric generation — all optional, all user-driven, never a black box in your render path.
Only in Pro Tools, where AAX requires PACE signing (a free iLok account is enough). VST3 in every other DAW needs no iLok at all.
A key arrives by email after purchase and activates offline inside the app — cryptographically signed, no account, no phone-home.
Windows 10/11 x64 today. macOS with AU support is on the v1.x roadmap.
One installer for every tier. Your license key unlocks Pro or Studio inside the app.
Download StemDesk for WindowsWindows 10/11, 64-bit. Every release ships with a SHA-256 checksum; the installer and plugins are code-signed. AAX use in Pro Tools requires a free iLok account. The optional local AI model is a one-time ~8 GB download, fetched only with your consent.