Topological fingerprints for audio identification

TL;DR

This paper introduces a topological audio fingerprinting method using persistent homology on mel-spectrograms, achieving robust and accurate identification of duplicate audio tracks even under distortions.

Contribution

The novel approach applies topological data analysis to audio signals, providing a new robust fingerprinting technique that outperforms existing methods under various distortions.

Findings

High accuracy in duplicate audio detection

Robustness to time stretching and pitch shifting

Outperforms existing fingerprinting methods

Abstract

We present a topological audio fingerprinting approach for robustly identifying duplicate audio tracks. Our method applies persistent homology on local spectral decompositions of audio signals, using filtered cubical complexes computed from mel-spectrograms. By encoding the audio content in terms of local Betti curves, our topological audio fingerprints enable accurate detection of time-aligned audio matchings. Experimental results demonstrate the accuracy of our algorithm in the detection of tracks with the same audio content, even when subjected to various obfuscations. Our approach outperforms existing methods in scenarios involving topological distortions, such as time stretching and pitch shifting.