Characterisation of composer style using high-level musical features

Characterisation of composer style using high-level musical features. Mearns, L., Tidhar, D., & Dixon, S. In Proceedings of 3rd International Workshop on Machine Learning and Music, of MML '10, pages 37–40, Firenze Italy, October, 2010. ACM.

Paper doi abstract bibtex

We describe a preliminary study looking into the characterisation of composer style. The primary motivation of the work is an exploration of methods to automatically extract high level, musicologically valid features. Such features facilitate machine-learning based stylistic classiﬁcation which, in contrast to previously published results, are more likely to yield musicological insights regarding style characteristics and compositorial techniques. We extract features from scores by Renaissance and Baroque composers, capturing their use of contrapuntal voice leading rules and musical intervals. A composer classiﬁcation task is performed to test the ability of the feature sets to characterise composer style, yielding an accuracy of 66%. We conclude that although the computation of higher level musical features is challenging, it can give useful insights into characteristics of style which are not revealed by lower level features.

@inproceedings{mearns2010characterisation,
	address = {Firenze Italy},
	series = {{MML} '10},
	title = {Characterisation of composer style using high-level musical features},
	isbn = {978-1-4503-0161-9},
	url = {https://dl.acm.org/doi/10.1145/1878003.1878016},
	doi = {10.1145/1878003.1878016},
	abstract = {We describe a preliminary study looking into the characterisation of composer style. The primary motivation of the work is an exploration of methods to automatically extract high level, musicologically valid features. Such features facilitate machine-learning based stylistic classiﬁcation which, in contrast to previously published results, are more likely to yield musicological insights regarding style characteristics and compositorial techniques. We extract features from scores by Renaissance and Baroque composers, capturing their use of contrapuntal voice leading rules and musical intervals. A composer classiﬁcation task is performed to test the ability of the feature sets to characterise composer style, yielding an accuracy of 66\%. We conclude that although the computation of higher level musical features is challenging, it can give useful insights into characteristics of style which are not revealed by lower level features.},
	language = {en},
	urldate = {2025-05-22},
	booktitle = {Proceedings of 3rd {International} {Workshop} on {Machine} {Learning} and {Music}},
	publisher = {ACM},
	author = {Mearns, Lesley and Tidhar, Dan and Dixon, Simon},
	month = oct,
	year = {2010},
	keywords = {Read},
	pages = {37--40},
}

Downloads: 0

{"_id":"gW95ff9M8Ncc8aorT","bibbaseid":"mearns-tidhar-dixon-characterisationofcomposerstyleusinghighlevelmusicalfeatures-2010","authorIDs":[],"author_short":["Mearns, L.","Tidhar, D.","Dixon, S."],"bibdata":{"bibtype":"inproceedings","type":"inproceedings","address":"Firenze Italy","series":"MML '10","title":"Characterisation of composer style using high-level musical features","isbn":"978-1-4503-0161-9","url":"https://dl.acm.org/doi/10.1145/1878003.1878016","doi":"10.1145/1878003.1878016","abstract":"We describe a preliminary study looking into the characterisation of composer style. The primary motivation of the work is an exploration of methods to automatically extract high level, musicologically valid features. Such features facilitate machine-learning based stylistic classiﬁcation which, in contrast to previously published results, are more likely to yield musicological insights regarding style characteristics and compositorial techniques. We extract features from scores by Renaissance and Baroque composers, capturing their use of contrapuntal voice leading rules and musical intervals. A composer classiﬁcation task is performed to test the ability of the feature sets to characterise composer style, yielding an accuracy of 66%. We conclude that although the computation of higher level musical features is challenging, it can give useful insights into characteristics of style which are not revealed by lower level features.","language":"en","urldate":"2025-05-22","booktitle":"Proceedings of 3rd International Workshop on Machine Learning and Music","publisher":"ACM","author":[{"propositions":[],"lastnames":["Mearns"],"firstnames":["Lesley"],"suffixes":[]},{"propositions":[],"lastnames":["Tidhar"],"firstnames":["Dan"],"suffixes":[]},{"propositions":[],"lastnames":["Dixon"],"firstnames":["Simon"],"suffixes":[]}],"month":"October","year":"2010","keywords":"Read","pages":"37–40","bibtex":"@inproceedings{mearns2010characterisation,\n\taddress = {Firenze Italy},\n\tseries = {{MML} '10},\n\ttitle = {Characterisation of composer style using high-level musical features},\n\tisbn = {978-1-4503-0161-9},\n\turl = {https://dl.acm.org/doi/10.1145/1878003.1878016},\n\tdoi = {10.1145/1878003.1878016},\n\tabstract = {We describe a preliminary study looking into the characterisation of composer style. The primary motivation of the work is an exploration of methods to automatically extract high level, musicologically valid features. Such features facilitate machine-learning based stylistic classiﬁcation which, in contrast to previously published results, are more likely to yield musicological insights regarding style characteristics and compositorial techniques. We extract features from scores by Renaissance and Baroque composers, capturing their use of contrapuntal voice leading rules and musical intervals. A composer classiﬁcation task is performed to test the ability of the feature sets to characterise composer style, yielding an accuracy of 66\\%. We conclude that although the computation of higher level musical features is challenging, it can give useful insights into characteristics of style which are not revealed by lower level features.},\n\tlanguage = {en},\n\turldate = {2025-05-22},\n\tbooktitle = {Proceedings of 3rd {International} {Workshop} on {Machine} {Learning} and {Music}},\n\tpublisher = {ACM},\n\tauthor = {Mearns, Lesley and Tidhar, Dan and Dixon, Simon},\n\tmonth = oct,\n\tyear = {2010},\n\tkeywords = {Read},\n\tpages = {37--40},\n}\n\n\n\n","author_short":["Mearns, L.","Tidhar, D.","Dixon, S."],"key":"mearns2010characterisation","id":"mearns2010characterisation","bibbaseid":"mearns-tidhar-dixon-characterisationofcomposerstyleusinghighlevelmusicalfeatures-2010","role":"author","urls":{"Paper":"https://dl.acm.org/doi/10.1145/1878003.1878016"},"keyword":["Read"],"metadata":{"authorlinks":{}},"downloads":0,"html":""},"bibtype":"inproceedings","biburl":"https://bibbase.org/zotero/fsimonetta","creationDate":"2019-04-22T13:04:36.110Z","downloads":0,"keywords":["read"],"search_terms":["characterisation","composer","style","using","high","level","musical","features","mearns","tidhar","dixon"],"title":"Characterisation of composer style using high-level musical features","year":2010,"dataSources":["RFLDZw5KyJdadDXDm","pzyFFGWvxG2bs63zP"]}