Observed arrival · 2026-09-24
Nearfield turns tracks into a similarity map
Nearfield lets you build a graph of music tracks and explore connections using audio-feature similarity.
○Open
⊠Login
$Paid
†Ads
✦Pretty
●Pro
◎Niche
◉Human
⚑Risk
ƒJS
- For
- Listeners mapping connections between tracks
- Worth noticing
- Its graph uses weighted groups of audio features, including 13 MFCCs, rather than the separate similarity score used for initial lookups.
Field notes
The page describes a graph built from track fingerprints, with cosine similarity determining its edges and distances. It assigns weights to four feature groups—MFCCs, spectral shape, tempo, and dynamics—and says analysis averages and varies measurements across two 45-second segments. Initial search and lookups go through the cosine.club API, whose separate similarity score is not used to draw the graph.
Observed signals
Read the marks
Editorial observations of this landing page, not a rating.
○OpenPublic substance visible
✦PrettyNotable craft visible
●ProPolished or operationally mature
◎NicheUnusually specific use
One card from the complete issue