TrackGleam browser mastering - looking for producer feedback

[trackgleam]

AUTO-ADMIN: Non-MP3, WAV, OGG, SoundCloud, YouTube, Vimeo, Twitter and Facebook links in this post have been protected automatically. Once the member reaches 5 posts the links will function as normal.

TrackGleam - private browser-based mastering

Hi everyone,

I have been working on TrackGleam, a browser-based mastering application, and would value feedback from producers here. Processing happens locally in the browser rather than uploading the track to a server.

Fast mastering workflow

• Multiple mastering styles, including streaming-ready, punchy, warm, bright, wide, loud, and natural
• Selectable intensity and loudness targets
• Streaming-ready WAV and MP3 export
• No account required for the standard workflow

Advanced analysis and controls

• Integrated LUFS and true-peak metering
• Loudness range and crest-factor analysis for dynamics
• Stereo-width measurement and mono-risk detection
• Artifact-risk analysis before export
• Tuned EQ with bass-heavy and punchy correction
• Fine control over tone, dynamics, width, and final level
• Reference-track matching for tone and loudness
• Original, mastered, and streaming-normalized comparison modes

These tools make TrackGleam useful beyond one-click mastering: you can inspect what is happening, compare versions, and shape the result before downloading it.

Try it
Open TrackGleam (https://trackgleam.com)

KVR product page
View TrackGleam on KVR

---

Feedback I would especially appreciate

• How well do the results translate across different genres?
• Do the advanced controls provide enough useful adjustment?
• How accurate are the loudness, stereo, and artifact-risk indicators on your material?
• Where does the reference matching need improvement?

Thanks for taking a look.

[Benedict]

I fed it an un-mastered wav of:
https://benedictroff-marsh.bandcamp.com ... ravel-time

Then compared:
Raw : Free Master : AI Master : Official Master (above)

The Raw was, of course, quiet so the Free Master, which turned up the levels, sounded better. Sweet really as it felt very natural, if not a bit quiet still compared to real-world expectations.

The AI Master was a bit louder still, and shinier overall. It still felt rather safe and "bloodless". It felt like it was 'true to what I sent' but not where I ended up. And I do not master to fix but to enhance what is there.

I then fed it my actual master wav to see what it said. It claimed a Crest of 13.5dB (I aim for 12 and my meters (which I trust) read under that, seeing I wanted it to feel full. Ok, meters are rarely that useful. They ping me for being -11.1 LUFS. I don't use LUFS. But even that is not the -8 LUFS commonly quoted as "what is needed". Choice I guess.

Their Free Master of my Master pulled levels back to solve their assumed problems, despite no real-world issues - if anything many would say I was too quiet; a deliberate choice so my material breathes, as that is what it is about. The AI Master then did its parade of processes and delivered something that, compared to my "big bully" Sene & Story feels twee and more like it is about flitting fairies than being held to a timetable (not to mention shades of Hi-NRG).

So, my 'professional' opinion: It did not make a technical mess - yay for not making it sound like Drunke as others have - but despite listing more processes/changes than I made, seemed to deliver less Story than my own master as it cannot understand the material and what it is looking to speak. Instead, it plays safe for 'technically correct' but 'bloodless' results, which in at least this case undid the Song itself.