TRACK INTELLIGENCE // BETA · SAMPLE DATA · POWERED BY PYXIS-1
See how the track moves, changes and holds together.
Through 70+ measurable audio characteristics, PYXIS-1 measures a song's rhythm, energy, harmony, timbre, spectral behaviour and structure, and turns these traits into one readable track profile.
Six sections in long phrases: intro, build, first drop, the longest break, the main drop and a mixable outro. The break → drop 02 transition is the track’s structural centre.
EXAMPLE INTERPRETATION FROM MEASURED OUTPUTS01 // THE PROFILE
One track. One readable profile.
The first level of analysis is a concise and professional summary, unbiased by charts.
A steady, club-oriented track with a long central breakdown, stable tempo and its highest sustained energy in the final drop. Measured values are read directly from the audio; predicted values are classification signals, shown with confidence.
EXAMPLE INTERPRETATION FROM MEASURED OUTPUTS02 // TIME & STRUCTURE
How the track unfolds
A structural score with the waveform, beat grid and sections on one timeline. Select a section to read its role.
The main drop begins after the longest breakdown and carries the track’s highest sustained energy, and brightness lifts with it. [example interpretation from measured outputs]
- DJSread the arrangement before it goes into a set
- PRODUCERScompare pacing and section balance against references
- SUPERVISORSlocate shifts in tension and intensity fast
- CATALOGUE TEAMSdescribe structure consistently across releases
03 // MOTION & DYNAMICS
How energy moves through the track
One continuous energy journey, with loudness and dynamic range layered beneath it, anchored to the structure.
Reading it: intensity climbs gradually, with no abrupt jumps; the long break releases energy before the strongest section; the final drop carries the global peak, then the outro steps down rather than cutting.
EXAMPLE INTERPRETATION FROM MEASURED OUTPUTSDJs check the track's flow before building a set. Producers study how arrangements are made. Curators focus on the pacing. Catalogue teams turn libraries into searchable tags. Our analysis just describes the music; the decision-making process is in your hands only.
04 // HARMONY & PITCH
Key and harmonic behaviour
Detected key, confidence and a readable chromagram strip up front, with full pitch evidence and harmonic relationships one step deeper.
The track stays centred on G minor throughout, with its strongest tonal movement in the break: pitch content spreads, then the final drop re-anchors the key.
05 // TIMBRE & SPECTRAL CHARACTER
The texture of the recording
A visual of the sound's frequencies, from deep bass to high air, described in plain language.
Texture read: dark and dense through the first half, smooth, sub-weighted and percussive. The final drop lifts brightness and opens the top end without losing weight.
EXAMPLE INTERPRETATION FROM MEASURED OUTPUTSDescribed as measured character — dark or bright, dense or open, smooth or percussive, stable or evolving.
06 // SONIC IDENTITY
A profile, not a label
Genre, style, mood and vibe as coordinated signals: stronger and weaker, never absolute.
Genre & style
detailed style: deep · dub-inflected · rolling
Mood & theme
Vibe & composition
What the analysis reads from the composition: instrument balance and phrasing, as shown in the dashboard.
Primary classification: deep house. Secondary signals suggest tech house and minimal influences. The profile keeps the overlap visible instead of forcing one answer.
07 // FROM DATA TO DECISION
One profile. Five professional readings.
For DJs
Read the track before you hear it — arrangement, tempo, key, movement and weight, at a glance.
- Structure: where every section starts and ends
- BPM: stable 124, confirmed across the beat grid
- Key: G minor · 6A, high confidence
- Energy journey: where it holds, lifts and releases
- Loudness and dynamics: how heavy it actually plays
Know the track before it goes anywhere near your set.
08 // TRUST
Measured, interpreted, never presented as absolute.
Classification depends on context. Styles overlap. Confidence is always shown, and every number is traceable back to how it was produced.
- Measured values (tempo, key, loudness, structure) are read straight from the audio: the same file always returns the same profile.
- Predicted values (genre, style, mood, vibe) are signals, shown with confidence. Secondary signals stay visible.
- Similar styles overlap; the profile presents the overlap instead of hiding it behind one label.
- Raw charts never stand alone; each carries a plain-language interpretation.
- Every result is traceable to its analysis run: engine version, date and scope.
09 // THE PLATFORM
The profile is just the beginning
Use the profile to find tracks that share its sonic character.
status: live in betaApply consistent analysis across a partner catalogue.
status: private pilotRetrieve structured track intelligence inside another product.
docs + guided pilotFOR MUSIC PROFESSIONALS
See your track as data you can use.
beta · analyse your own track and read the full profileFOR CATALOGUES & PLATFORMS