Robust Quad-Based Audio Fingerprinting

8 years 11 days ago

Download www.cp.jku.at

—We propose an audio ﬁngerprinting method that adapts ﬁndings from the ﬁeld of blind astrometry to deﬁne simple, efﬁciently representable characteristic feature combinations called quads. Based on these, an audio identiﬁcation algorithm is described that is robust to noise and severe time-frequency scale distortions and accurately identiﬁes the underlying scale transform factors. The low number and compact representation of content features allows for efﬁcient application of exact ﬁxedradius near-neighbour search methods for ﬁngerprint matching in large audio collections. We demonstrate the practicability of the method on a collection of 100,000 songs, analyse its performance for a diverse set of noise as well as severe speed, tempo and pitch scale modiﬁcations, and identify a number of advantages of our method over two state-of-the-art distortionrobust audio identiﬁcation algorithms.

Reinhard Sonnleitner, Gerhard Widmer

Real-time Traffic