Recent masters

Reopen your retained results without mastering again. Albums and multi-song batches: one hour after completion. Single masters: five focused minutes, with a one-hour RAM limit. Save your files locally.

VERIFY TO ACTIVATE

Activate temporary Free access

Access details

Verify your email for all models, reference matching, up to 24 tracks per batch and 5,000 weekly credits. Each model uses the credits shown before rendering.

Access & credits

  1. UploadYour audio files
  2. AnalysisSource & workflow
  3. SettingsModel & finish
  4. Listen & downloadYour finished master

Step 1 of 4: Upload your audio.

Choose audio file(s)

WAV, AIFF, FLAC or MP3. Verified full access: up to 24 tracks. Total upload: 250 MB.

Select File(s)

Sign In Required

An OrynosMusic account is required to access the engine.

Acoustic Analysis

Select Workflow

Orynos AI Mastering

Choose your Orynos AI model. It handles tone, dynamics and the final loudness for you.

Manual controls

Set the tone, stereo width and loudness. Available tools follow the model you choose.

Mud Cleaner

Reduce sub-bass rumble and low-mid boxiness. 10 credits per track.

Reference Match Included

Use a reference track to guide tonal balance and loudness.

Shape your master.

Choose your model and analysis depth, then adjust only what your track needs.

Optional, off by default. Originals stay in RAM only, within this result's existing download window and at most one hour after completion. Each revision renders from the original with the same model and depth, using normal credits for every track. Download your files before they expire.

SUN · GALAXY · AURORA

Binaural Spatial Master

Build an additional headphone mix with width, height cues and a room around your stereo master.

Available with Sun, Galaxy, Aurora and Blackhole.

Shape the headphone space

Short room arrivals for depth.

A softer, less pinpoint room field.

Keep the center image more anchored.

Keep low frequencies more stable.

Reduce room blur around sharp attacks.

Include the derived scene and metadata.

Your stereo master stays unchanged. Headphones recommended; spatial impression varies. A stereo source cannot reveal isolated instruments or your individual hearing profile.

Orynos AI · distinct DSP models

Choose your model.

Access & credits ↗

Loading model access…

Analysis depth

One final master. Higher depths use more source measurements and finer output checks; the multiplier describes analysis budget, not a sound-quality score.

Frame-by-frame narrow resonance control; correction capped at 3.5 dB, with onset protection.

Uses spectral onset motion to retain attacks during resonance control.

Reduces opposed stereo energy only in affected frequency/time cells.

Deshimmer-derived narrow high-frequency artifact suppression. Protects transients and broad musical events. Audition: musical tones can also trigger it.

DSP source ↗

Independent bass, mid and high-band gain control.

Reduce low-frequency side energy below 110 Hz.

Measure the mix and apply bounded five-band tonal correction.

Gentle broadband gain control keeps both channels linked.

Oversampled nonlinear coloration and extra low-frequency body.

Finishing collection · 7 optional tools

Adaptive control of moving sub-bass.

A gentle lift to attack with less sustain.

Soften the upper spectrum with a restrained high-band blend.

Remove energy below 30 Hz before model processing.

Smooth the final 50 ms to reduce end clicks.

Reduce side energy when channels have negative correlation.

Level-dependent attenuation of harsh upper-band peaks.

Create a custom profile Included
Discord collection Members

Connect Discord in Settings to unlock three exclusive recipes and a loudness-matched listening pack.

Connection settings ↗

Choose a model to see its processing path and controls.

All modules and models are included for verified accounts during the temporary Free phase. Access & credits

Your manual controls. Set loudness and tone before you render. Model-specific tools follow the selected signal path.

Target LUFS
?
LUFS: Loudness Units relative to Full Scale. -14.0 is optimized for Spotify/Apple Music, while -9.0 is for heavy club tracks.

Output loudness target. Louder settings can reduce dynamics; peak protection may limit the final level.

-14.0
AI Mastering Governor
Controls how strongly the AI may process the song and how much macro-dynamics it must preserve. These are independent: Strong + Preserve is supported.
Active
Requested
-14.0
AI Recommended
After render
Max Safe
Solver
Achieved
Measured

Acoustic Profile
?
Adapts dynamic curves based on specific presets. EDM focuses on transient punch, Glue creates cohesiveness.

Baseline profile preset.

True Peak Ceiling
?
Prevents inter-sample peak distortion. -1.0 dB is standard for Spotify conversion safe headroom.

Peak protection for the final output. Leave headroom when the master will be encoded or streamed.

Bit Depth
?
24-bit is modern high-fidelity lossless standard. 16-bit applies precise dither.

Export bit resolution.

Bass Boost

Low-shelf shelf adjustment centered at 120Hz.

0.0 dB

Mid Presence

High-Q bell curves adjustments at 2.5kHz range.

0.0 dB

Highs & Air

Adds high-frequency breath and clarity at 8kHz+.

0.0 dB

Saturation

Even/odd analog harmonics warmth intensity.

0.0

Stereo Width

1.00 keeps the source width. Lower values narrow it; higher values add width. Check mono compatibility before downloading.

1.00

Spectral Mud Optimizer

Intelligently suppresses muddy low-mids in real time.

Smart De-Esser

Suppress vocal harshness sibilants surgically.

Smart Voice Focus

Carves transient instrumental space to spotlight the vocals.

Dynamic Clarity

Intelligently eliminates harsh boxy resonances.

Punch Enhancer

Amplifies drum transient attacks transparently.

Air & Shine

Applies smooth high-end shine to polish the mix.

Analog VCA Glue Compressor

Tightens and solidifies cohesive drum presence.

Bass Monoizer

Forces sub frequencies under 90Hz to phase-safe mono.

Vintage Tape Emulation

Simulates analog wow, flutter, and soft tube tape compression.

Hi-Fi 192kHz Upscaler

HQ upsampling rendering to eliminate internal DSP aliasing.

Analog Console Crosstalk

SSL/Neve nonlinear stereo leakage bleed emulation.

Multiband Transient Shaper

Applies transient shaping to split frequency bands.

Phase Rotation

Fixed asymmetric stereo waveforms to gain headroom.

DC Offset Removal

Filters out dynamic sub-sonic offset rumble.

Dynamic Upward Expansion

Re-expands peak dynamics in squashy recordings.

Mid/Side Auto-Balancer

Adjusts side-channel gain using the measured mid/side RMS balance.

Tube Exciter

Generates harmonic tubes shine presence on mids.

Pre-Chain Auto-Leveler (AGC)

Correct gain stages input files precisely to -18dBFS.

16x Oversampled ISP Limiter

Measures oversampled peaks and limits them toward your selected ceiling.

Soft Clipper

Rounds peaks with oversampled soft saturation for controlled harmonics.

Centre-band Dynamics

Controls centre-channel level using a 300-4,000 Hz envelope detector.

Auto-Phase Aligner

Align phase relationships across sub-bass ranges.

Cinematic Micro-Reverb

Psychoacoustic three-dimensional micro-room depth.

AI Resonance Suppressor

Surgically tames sharp resonances in real-time.

Vocal Rider AI

Auto rides dynamic vocal level to sit perfectly on beat.

Lo-Fi Artifact Generator

Adds warm crackle hum tape hiss texturizer.

5 INCLUDED TOOLS

Finish with confidence.

Included tools, plus Sun-only frequency-layer export. Enabled tools add measured results and files to your release ZIP.

Stereo Balance Repair Included

Lower the stronger channel before mastering to correct accidental imbalance. Intentional panning can also change; the report records the correction.

Loudness Timeline Included

Find loudness jumps and quiet sections in the finished master. Export a chart, CSV and JSON with LUFS in 3-second windows and sample peaks.

Normalization Audition Included

Hear the first 30 seconds after loudness normalization. Measures the entire master and only turns it down; the original master stays intact.

Crossfade Loop Export Included

Create an additional loop WAV with a blended end-to-start seam for games, installations and samples. The loop is rotated and shortened by the crossfade; choose your musical cut before uploading.

Frequency Layer Export SUN · GALAXY

Export complementary low, mid and high WAV layers for inspection and creative remixing. The included report explains how to reconstruct the master. These are frequency bands, not instrument stems.

Your master is ready.

Processed in 0.0s

OrynosMusic Master
OrynosMusic DSP
0:00 0:00

Translation Check

Linked A/B listening

Finish a master to check how it translates. Downloads stay unchanged.

Listening preview only. No extra credits. Small speaker is a band-limited check, not a simulation of a specific device.

0:00.00 – 0:10.00

Focus the player for Space play / pause · A original · B master · L loop.

Join Discord

How to master a song online

Prepare your mix ↗
  1. Upload your finished mix.

    Start with a stereo export that has no clipping. Upload once; analysis and mastering reuse the same audio.

  2. Choose a mastering model.

    Use automatic mastering for a simple workflow, or adjust the available controls. Review the measured recommendations for your track.

  3. Listen, compare, download.

    Compare the original and master at matched volume. Check quiet passages and loud sections before choosing your download format.

Which audio formats can I use?

Upload WAV, AIFF, FLAC, MP3 or M4A. Choose a lossless source when available; converting an MP3 to WAV will not restore lost detail. Finished masters can be downloaded as WAV, FLAC or MP3. At 192 kHz, choose WAV or FLAC.

Which model and analysis mode should I choose?

Models have different processing paths, controls and credit costs. Start with Standard and listen before increasing the analysis budget. Power, Ultra and Max deepen the source analysis on compatible models; they produce one final master. Compare model and mode choices.

How long is my finished master available?

You have five minutes of focused result-tab time to download. The timer pauses when you leave the tab; audio is removed from RAM no later than one hour after completion. Upload, queueing and processing do not use this time. Save the files you want to keep.