Predicting Expressive Dynamics in Piano Performances using Neural Networks

Sam van Herwaarden, Maarten Grachten, W. Bas de Haas

    Research output: Chapter in Book/Report/Conference proceedingConference contributionAcademicpeer-review

    Abstract

    This paper presents a model for predicting expressive
    accentuation in piano performances with neural networks.
    Using Restricted Boltzmann Machines (RBMs), features
    are learned from performance data, after which these
    features are used to predict performed loudness. During
    feature learning, data describing more than 6000 musical
    pieces is used; when training for prediction, two datasets
    are used, both recorded on a Bosendorfer piano (accurately
    measuring note on- and offset times and velocity values),
    but describing different compositions performed by
    different pianists. The resulting model is tested by predicting
    note velocity for unseen performances. Our approach
    differs from earlier work in a number of ways: (1) an
    additional input representation based on a local history of
    velocity values is used, (2) the RBMs are trained to
    result in a network with sparse activations, (3) network
    connectivity is increased by adding skip-connections, and (4)
    more data is used for training. These modifications result
    in a network performing better than the state-of-the-art on
    the same data and more descriptive features, which can be
    used for rendering performances, or for gaining insight into
    which aspects of a musical piece influence its performance.
    Original languageEnglish
    Title of host publicationProceedings of the 15th Conference of the International Society for Music Information Retrieval (ISMIR 2014)
    Subtitle of host publicationOctober 27 - 31, 2014 Taipei, Taiwan
    EditorsHsin-Min Wang , Yi-Hsuan Yang , Jin Ha Lee
    PublisherInternational Society for Music Information Retrieval
    Pages45-52
    Number of pages6
    Publication statusPublished - 2014
    EventInternational Society for Music Information Retrieval Conference - Taipei, Taiwan, Province of China
    Duration: 27 Oct 201431 Oct 2014

    Conference

    ConferenceInternational Society for Music Information Retrieval Conference
    Country/TerritoryTaiwan, Province of China
    CityTaipei
    Period27/10/1431/10/14

    Fingerprint

    Dive into the research topics of 'Predicting Expressive Dynamics in Piano Performances using Neural Networks'. Together they form a unique fingerprint.

    Cite this