Guides
What generated tracks get wrong, measured, and how to fix each one. Written from the numbers, not the marketing.
Fixing and finishing AI tracks
- Suno tracks sound hissy or metallic. Here is what is actually there.
Hiss, a metallic ring in the mids, and a boxed vocal are the three most common Suno v5 artifacts. What they measure as, and how to fix each without dulling the track.
- Why AI songs stop at 16 kHz, and why an EQ boost cannot fix it
AI music generators stop the spectrum near 16 kHz. Here is why that makes a track sound dull, why an EQ boost cannot fix it, and what a real rebuild does instead.
- How loud should my song be for Spotify, Apple Music and YouTube
Spotify normalizes to -14 LUFS, Apple Music to -16, YouTube to about -14. What that means for your master, what loudness to aim for by genre, and the one number that matters more than loudness.
- LUFS, true peak and crest factor, explained for people who just want the song to sound right
The three numbers on every mastering report, what each one means for how your song sounds, and the ranges a finished track usually sits in.
- Suno to DistroKid: a release checklist that survives review
The steps between a Suno export and a live Spotify release: file format, loudness, true peak, artwork, AI disclosure, and the mistakes that get a release rejected.
- Udio disabled downloads. What to do with the songs you already have
Udio turned off downloads across its tiers during its label licensing rollout. If you still hold exports, here is how to finish and release them, and what to check first.
- The 350 Hz problem: why AI tracks sound boxy, and how to clean the low mids
A bump between 250 and 450 Hz is the most common reason a generated track sounds boxy and does not translate off headphones. How to find it and how much to cut.
- Why AI songs sound narrow, and how to widen without ruining mono
Generated tracks usually have a side signal 15 to 20 dB under the mid. What that means, how to widen it safely, and why the bass must stay in the center.
- Mastering on an iPhone without uploading anything
Every online mastering service uploads your unreleased track to a server. Burnish for iOS runs the whole chain on the phone, offline, and works as an AUv3 plugin in GarageBand. What that changes.
- AI mastering without a subscription: what you actually pay per song
Most mastering services charge monthly whether you release or not, and some keep a royalty after you cancel. Here is what a song really costs on each model, and why Burnish charges once.
- Mastering to a reference track: what it can and cannot copy
A reference master can give you a target for tone, width and loudness. It cannot give you the arrangement or the mix. What reference matching does, how much to apply, and when to skip it.
- AI disclosure in 2026: what to fill in, and what a mastering tool should never do
Distributors now ask whether a track used AI, Spotify credits AI contributions, and provenance tags are arriving in audio. What to fill in, what happens if you do not, and why detection evasion is a trap.
- Making an EP sound like one record: consistent loudness across tracks
Mastering songs one at a time gives you an album where track three is louder than track two. How album mastering sets one loudness and one tone across a release, and what to do about the quiet ballad.
- Bass mono: why club systems and vinyl need it
Stereo information below about 100 Hz cancels on club systems and cannot be cut to vinyl. What bass mono does, where to set the crossover, and when to leave the low end stereo.
- Harsh S sounds on AI vocals: de-essing without lisping
Generated vocals often carry sharp sibilance between 5 and 9 kHz. How a de-esser works, where the S lives, how much to take, and how to avoid the lisp that comes from taking too much.
- Why your master distorts after upload: true peak and encoder overshoot
The file was clean in your player and crackles on Spotify. That is true peak. Why lossy encoders push peaks past zero, how to measure it properly, and the ceiling that fixes it.
- Mastering inside GarageBand on iPad with an AUv3 plugin
How to put a mastering chain on the master track in GarageBand or Logic Pro for iPad using an AUv3 plugin, what the two controls do, and how to bounce a finished file.
- Suno v5 artifacts, measured
We measured recent Suno exports against commercial references. The cutoff frequency, the low-mid excess, the resonances, the width and the loudness, with numbers, and what each one takes to fix.
- Make a before and after video of your master for TikTok, Reels and Shorts
The clip that shows a track getting fixed is the most shared thing a mastering tool can produce. How to make one that is honest, level-matched, and worth watching for fifteen seconds.
- Free AI mastering tools vs Burnish: what free costs you
There are good free mastering tools. Here is what they do, what they skip, where your file goes, and the cases where paying $5 for a master is the cheaper option.
By genre: the numbers each profile uses
Generated from the engine's own profiles. Loudness targets, bass mono crossover, width, crest, the signature EQ, and the compression pressure for each genre.
- Mastering AI-made Acoustic / Folk / Singer-Songwriter tracks: the numbers Burnish uses
- Mastering AI-made Ambient / Downtempo tracks: the numbers Burnish uses
- Mastering AI-made Classical / Orchestral tracks: the numbers Burnish uses
- Mastering AI-made Country / Americana tracks: the numbers Burnish uses
- Mastering AI-made Deep House tracks: the numbers Burnish uses
- Mastering AI-made Drum & Bass tracks: the numbers Burnish uses
- Mastering AI-made EDM / Big Room tracks: the numbers Burnish uses
- Mastering AI-made Funk / Soul tracks: the numbers Burnish uses
- Mastering AI-made Hip-Hop / Trap tracks: the numbers Burnish uses
- Mastering AI-made House / Tech House tracks: the numbers Burnish uses
- Mastering AI-made Jazz tracks: the numbers Burnish uses
- Mastering AI-made Reggaeton / Latin tracks: the numbers Burnish uses
- Mastering AI-made Metal tracks: the numbers Burnish uses
- Mastering AI-made Pop tracks: the numbers Burnish uses
- Mastering AI-made Reggae / Dancehall tracks: the numbers Burnish uses
- Mastering AI-made R&B / Soul tracks: the numbers Burnish uses
- Mastering AI-made Rock / Indie tracks: the numbers Burnish uses
- Mastering AI-made Techno tracks: the numbers Burnish uses