Sound Fingerprinter
Play a clip and see how audio fingerprinting picks out a song from its spectrogram peaks.
Coming soon
This tool is being built. It will run right here, in your browser.
How it works
Method
The clip is turned into a spectrogram, its strongest peaks are paired into hashes, and the hashes are matched against a library.
Data
A library of openly licensed music tracks; your microphone audio stays in your browser.
Model
Constellation-map hashing as described in the original Shazam paper.
Build
Web Audio analysis in the browser with a pre-built hash index.