Model-based cover song detection via threshold autoregressive forecasts

TitleModel-based cover song detection via threshold autoregressive forecasts
Publication TypeConference Paper
Year of Publication2010
Conference NameACM Multimedia, Int. Workshop on Machine Learning and Music (MML)
AuthorsSerrà, J., Kantz H., & Andrzejak R. G.
Pagination13-16
Conference Start Date25/10/2010
PublisherACM
Conference LocationFirenze, Italy
ISBN Number978-1-60558-933-6
Keywordscover songs, Information retrieval, Music, prediction, threshold autoregressive models
AbstractCurrent systems for cover song detection are based on a model-free approach: they basically search for similarities in descriptor time series reflecting the evolution of tonal information in a musical piece. In this contribution we propose the use of a model-based approach. In particular, we explore threshold autoregressive models and the concept of cross-prediction error, i.e. a measure of to which extent a model trained on one song's descriptor time series is able to predict the covers'. Results indicate that the considered approach can provide competitive accuracies while being considerably fast and with potentially less storage requirements. Furthermore, the approach is parameter-free from the user's perspective, what provides a robust and straightforward application of it.
preprint/postprint documentfiles/publications/mml0823-serra.pdf
DOI of final publication10.1145/1878003.1878008
intranet