Repetition is a fundamental element in generating and perceiving structure. In audio, mixtures can often be understood as a background component that is generally more repeating in time, superimposed with a foreground component that is generally more variable in time (e.g., a song with varying vocals overlaid on a repeating accompaniment, or a recording with a varying speech mixed up with a repeating noise). On this basis, Zafar Rafii presents the REpeating Pattern Extraction Technique (REPET), a novel and intuitive approach for separating the repeating background from the non-repeating foreground in an audio mixture. The basic idea is to identify repeating elements in the mixture by measuring self-similarity along time, derive repeating models by averaging the repeating elements over their repetitions, and extract the repeating structure by comparing the repeating models to the mixture. Unlike other source separation approaches, REPET does not depend on special parametrizations, does not rely on complex frameworks, and does not require external information. Because it is only based on repetition, it has the advantage of being simple, fast, blind, and therefore completely and easily automatable.

Zafar Rafii is currently a research engineer at Gracenote. He received a Master of Science in Electrical Engineering from both Ecole Nationale Superieure de l’Electronique et de ses Applications (ENSEA) in France and Illinois Institute of Technology (IIT) in Chicago, and a PhD in Computer Science from Northwestern University. He also worked as a research engineer at Mist Technologies, now known as Audionamix, in France and as a PhD intern at Gracenote. His research interests are centered on audio analysis, at the intersection of signal processing, machine learning, and cognitive science. Some of the projects he has been working on are, among others, music genre classification, adaptive user interfaces, mono and stereo source separation, speech enhancement, audio fingerprinting and cover Identification. More information about his work, including publications, codes, and examples, can be found on his website: http://zafarrafii.com/

Presented by MIR @ Berkeley