The long version
How the scoring actually works.
The homepage gives you the short answer: we score your take against the ceiling of the gear that recorded it. This is the rest of it, for anyone who wants to know what the engine measures before they trust a number it hands them.
From a raw take to a fix, in one pass.
No dashboards to decode. You get a read you can act on.
Analysis
Upload a clip. We measure noise floor, sibilance, room, levels, and clipping. Same engine on every plan.
Diagnosis
Two scores: how well you used your gear, and how far that gear can go. Plus the one thing costing you the most points.
Fix
Fixes ordered by impact, not a wall of metrics. Ask Yemi why, or clean up the take right there.
Why it's honest
We grade you against your ceiling, not a studio's.
Performance says how close your take came to your gear's best. Ceiling says how good that best is. Keeping them separate cuts both ways: a clean take on a cheap mic finally reads as what it is, and a rough take can't hide behind an expensive one.
And when your gear isn't the problem, the report says exactly that. Nobody gets upsold on a mic they don't need.
first fix: gain staging (peaks at -1 dBFS)
The publish kit
Everything you need to post. Not just an MP4.
Most editors stop at “render.” We hand you the whole upload: the file, the chapters, the captions, and a final read on your sound.
Export presets by destination
YouTube 1080p / 2K / 4K, Shorts · Reels · TikTok, or Square. One tap sets the resolution and aspect. No codec menus.
YouTube chapters, validated
Drop chapter markers and copy a ready-to-paste list. We check YouTube's silent rules (0:00 first, 3+ chapters, 10s apart) so it won't be quietly ignored.
Captions as real files
SRT, VTT, or a plain transcript. Each one is timed to match the video exactly, so viewers can toggle them and search can index them.
A loudness verdict at export
“Your audio: −16.2 LUFS. Ready for YouTube / Spotify.” The moat, one last time, right when you export.