A visualizer that knows the difference between a verse and a drop
Most visualizers map loudness to size and stop there, so a quiet intro looks like a drop. Galaxy reads the arrangement — which instrument is carrying the track, where the phrase is, and whether this is a build, a drop or a breakdown.
Who this is for
- Producers tired of visuals that ignore the arrangement
- Anyone making a full-length video
- People who care that the picture means something
Which voice is leading
At any moment one thing carries a track — the bass line, a vocal, the hats, the drums. Galaxy works that out continuously and gives the leading voice the foreground, holding the rest back. That is why the picture reads instead of twitching.
Phrases, not frames
Music arrives in fours and eights. Camera moves and scene changes resolve on those boundaries rather than drifting at random, so the visual feels edited rather than generated.
Restraint, on purpose
There is a budget for how much of a scene may move at all. In an intro or a breakdown it drops to near stillness, so the loud moments actually land. Measured on a test track, the budget runs about 0.11 in an intro and 0.73 in a drop.
How to do it
- Open Galaxy — it runs in the browser on WebGL
- Choose a song or play the demo
- Pick a cosmic scene and add your logo or title
- Record, or export an MP4 at 1080p–4K
Questions people ask about visualizer that follows song structure
Does it need me to tag the sections?
No. It is detected live from the audio, with no upload and no analysis service.
What if my track has no obvious drop?
Then no drop fires. The visual follows what is actually there rather than inventing a moment.
Does the 2D app do this too?
BLOOB maps song structure for builds and drops and steers scene changes from it.
Is it exact?
Sections are detected, not known. Intro, groove, drop and breakdown are reliable; a build is the hardest and is sometimes read as groove.
Related: galaxy music visualizer · cinematic music visualizer · beat visualizer · three js music visualizer