AudioVibes
Dashboard Runs on your machine build 2026-09-05.32

Your sessions

Drop in a master. Your audio never leaves your machine. It is decoded and analyzed in your browser, and it is never transmitted. There is no size limit.

History

No sessions yet
A/BTrack ScoreDelivery Analyzed

Nothing analyzed yet. Add a master on the right and press Process.

New analysis

Master WAV · FLAC · MP3
Drop a file here, or
Any size. Your audio never leaves your machine.
Add a master to begin
This is the real engine. The same measurements as the Python core — BS.1770 loudness, EBU R128 loudness range, 4× oversampled true peak, stereo phase, spectral tilt — ported to JavaScript. On the test fixtures its integrated loudness agrees with pyloudnorm, an independent BS.1770 implementation, to within 0.05 LUFS, and every score it shares with the Python core agrees to within 0.5 points. Those loudness and peak figures are measured against published standards; the genre profiles the scores are weighed against are declared starting points, not calibrated against a corpus of masters. Tempo, key and structure still need the server engine and are not shown here.

Analyzing

Engine
AudioVibes v0.1.0 · in-browser
Elapsed
0.0s
Starting…0%
Real work, not a spinner. The analysis runs in a background worker so this page stays responsive. On a five-minute master expect roughly 15–25 seconds depending on your machine — the true-peak pass, which interpolates four points between every sample, is most of it.
0.0
/ 100

Saved clips are held in your browser's own storage (IndexedDB) — up to six clips within a 420 MB budget, oldest evicted first. Browser storage is not encrypted: anyone using that browser profile can replay them.

Findings

Category scores

weighted into the headline

Octave balance

tonal profile

Loudness over time

short-term, 3 s window

Key measurements

Source

Not measured here

Tempo, key, structure and performance need librosa and run on the server engine. Harmonic scoring is withheld on any full mix regardless — chroma saturates on a dense master, so it needs stems to mean anything.

← Back to dashboard
Analyze another track
Difference

Listen A / B

0:00
0:00
0.0 dB

What moved

largest change first

Category scores

Findings

Sources

← Back to dashboard