What is Audio Fingerprinting?
Also called: acoustic fingerprint
Audio fingerprinting is creating a compact digital signature from the sound of a recording, so the same recording can be recognized later, even from a short or noisy clip.
A fingerprint is built from robust features of the audio, typically the pattern of strong peaks in a spectrogram over time, rather than from the file's data. That is why it still matches after the audio is compressed, played through speakers or mixed with crowd noise.
It powers song recognition apps, identifying tracks in DJ mixes, broadcast monitoring for royalties, and copyright systems such as YouTube's Content ID.
A fingerprint identifies a specific recording. A cover version or a live performance of the same song usually needs other methods to be matched.
Put it to work
Identify the tracks in a DJ mix.
Identify Tracks in DJ Mix