Computational Analysis of the Live Music Archive

Sean Bechhofer, Simon Dixon, George Fazekas, Thomas Wilmering, Kevin Page

    Research output: Contribution to conferencePoster

    70 Downloads (Pure)

    Abstract

    The Computational Analysis of the Live Music Archive (CALMA) project aims to facilitate investigation and schol- arship related to live music through development of a Link- ed Data service combining metadata captured during depo- sition of audio to the Internet Archive, with computational analyses over these recordings through feature extraction, clustering, and classification. In this poster and demon- strator we introduce the architecture, tools, and data struc- tures we have developed to create this combined resource, and provide a first release of the dataset including prove- nance metadata to assist its interrogation and reuse. We also show the early results of questions assessed over the data that (i) aid resolution of uncertain metadata, identifi- cation of potential errors, and validation of existing entries, and (ii) provide metrics for broad patterns in performance variation that can be used to select subsets within the data for further longitudinal and musicological study.
    Original languageEnglish
    Publication statusPublished - 2014
    EventISMIR2014, 15th International Society for Music Information Retrieval Conference - Tapei, Taiwan
    Duration: 1 Jan 1824 → …

    Conference

    ConferenceISMIR2014, 15th International Society for Music Information Retrieval Conference
    CityTapei, Taiwan
    Period1/01/24 → …

    Keywords

    • music "information retrieval" "linked data"

    Fingerprint

    Dive into the research topics of 'Computational Analysis of the Live Music Archive'. Together they form a unique fingerprint.

    Cite this