Cognitive Features for Cover Song Retrieval and Analysis

    Research output: Contribution to conferenceAbstractOther research output


    This article presents the first results of a study on large-scale automatic identification of derivative works in early popular music. We present a music cognition-inspired approach to (audio) version detection that is (1) tailored for early 20th century recordings, (2) based on simple, index- able feature representations and (3) allows for interpretation of the resulting musical descriptions. For this purpose, two new descriptor types are introduced: pitch bi-histograms and chroma correlation coefficients.
    Original languageEnglish
    Publication statusPublished - Jun 2014
    EventInternational Workshop on Folk Music Analysis - Istanbul, Turkey
    Duration: 12 Jun 201413 Jun 2014


    ConferenceInternational Workshop on Folk Music Analysis


    Dive into the research topics of 'Cognitive Features for Cover Song Retrieval and Analysis'. Together they form a unique fingerprint.

    Cite this