RightSignal
1 of 4

Your RightSignal

Concept of the day1 / 3

Audio fingerprinting

The concept

How Shazam names a song from a few seconds of noisy audio

The app never compares raw sound. It turns your clip into a spectrogram, a picture of which pitches (frequencies) are loud at each moment, and keeps only the brightest spots, called peaks.

It then pairs nearby peaks and turns each pair into a short code: first pitch, second pitch, and the time gap between them. Those codes are the song's fingerprint, and a ten-second clip produces many of them.