Improving Voice Separation by Better Connecting Contigs. Guiomard-Kagan, N., Giraud, M., Groult, R., & Levé, F. In Mandel, M. I., Devaney, J., Turnbull, D., & Tzanetakis, G., editors, Proceedings of the 17\textsuperscriptth International Society for Music Information Retrieval Conference, ISMIR 2016, New York City, United States, August 7-11, 2016, pages 164–170, 2016. tex.date-added: 2019-03-29 16:18:17 +0100 tex.date-modified: 2019-03-29 16:18:17 +0100![Improving Voice Separation by Better Connecting Contigs [pdf]](https://bibbase.org/img/filetypes/pdf.svg "pdf")
Paper abstract bibtex Separating a polyphonic symbolic score into monophonic voices or streams helps to understand the music and may simplify further pattern matching. One of the best ways to compute this separation, as proposed by Chew and Wu in 2005 [2], is to first identify contigs that are portions of the music score with a constant number of voices, then to progressively connect these contigs. This raises two ques-tions: Which contigs should be connected first? And, how should these two contigs be connected? Here we propose to answer simultaneously these two questions by consid-ering a set of musical features that measures the quality of any connection. The coefficients weighting the features are optimized through a genetic algorithm. We benchmark the resulting connection policy on corpora containing fugues of the Well-Tempered Clavier by J. S. Bach as well as on string quartets, and we compare it against previously pro-posed policies [2, 9]. The contig connection is improved, particularly when one takes into account the whole content of voice fragments to assess the quality of their possible connection.
@inproceedings{guiomard-kagan_improving_2016,
title = {Improving {Voice} {Separation} by {Better} {Connecting} {Contigs}},
url = {https://wp.nyu.edu/ismir2016/wp-content/uploads/sites/2294/2016/07/129 \_Paper.pdf},
abstract = {Separating a polyphonic symbolic score into monophonic voices or streams helps to understand the music and may simplify further pattern matching. One of the best ways to compute this separation, as proposed by Chew and Wu in 2005 [2], is to first identify contigs that are portions of the music score with a constant number of voices, then to progressively connect these contigs. This raises two ques-tions: Which contigs should be connected first? And, how should these two contigs be connected? Here we propose to answer simultaneously these two questions by consid-ering a set of musical features that measures the quality of any connection. The coefficients weighting the features are optimized through a genetic algorithm. We benchmark the resulting connection policy on corpora containing fugues of the Well-Tempered Clavier by J. S. Bach as well as on string quartets, and we compare it against previously pro-posed policies [2, 9]. The contig connection is improved, particularly when one takes into account the whole content of voice fragments to assess the quality of their possible connection.},
booktitle = {Proceedings of the 17{\textbackslash}textsuperscriptth {International} {Society} for {Music} {Information} {Retrieval} {Conference}, {ISMIR} 2016, {New} {York} {City}, {United} {States}, {August} 7-11, 2016},
author = {Guiomard-Kagan, Nicolas and Giraud, Mathieu and Groult, Richard and Levé, Florence},
editor = {Mandel, Michael I. and Devaney, Johanna and Turnbull, Douglas and Tzanetakis, George},
year = {2016},
note = {tex.date-added: 2019-03-29 16:18:17 +0100
tex.date-modified: 2019-03-29 16:18:17 +0100},
keywords = {\#nosource, ⛔ No DOI found},
pages = {164--170},
}
Downloads: 0
{"_id":"2dDbm6oGoGeZTez6W","bibbaseid":"guiomardkagan-giraud-groult-lev-improvingvoiceseparationbybetterconnectingcontigs-2016","author_short":["Guiomard-Kagan, N.","Giraud, M.","Groult, R.","Levé, F."],"bibdata":{"bibtype":"inproceedings","type":"inproceedings","title":"Improving Voice Separation by Better Connecting Contigs","url":"https://wp.nyu.edu/ismir2016/wp-content/uploads/sites/2294/2016/07/129 \\_Paper.pdf","abstract":"Separating a polyphonic symbolic score into monophonic voices or streams helps to understand the music and may simplify further pattern matching. One of the best ways to compute this separation, as proposed by Chew and Wu in 2005 [2], is to first identify contigs that are portions of the music score with a constant number of voices, then to progressively connect these contigs. This raises two ques-tions: Which contigs should be connected first? And, how should these two contigs be connected? Here we propose to answer simultaneously these two questions by consid-ering a set of musical features that measures the quality of any connection. The coefficients weighting the features are optimized through a genetic algorithm. We benchmark the resulting connection policy on corpora containing fugues of the Well-Tempered Clavier by J. S. Bach as well as on string quartets, and we compare it against previously pro-posed policies [2, 9]. The contig connection is improved, particularly when one takes into account the whole content of voice fragments to assess the quality of their possible connection.","booktitle":"Proceedings of the 17\\textsuperscriptth International Society for Music Information Retrieval Conference, ISMIR 2016, New York City, United States, August 7-11, 2016","author":[{"propositions":[],"lastnames":["Guiomard-Kagan"],"firstnames":["Nicolas"],"suffixes":[]},{"propositions":[],"lastnames":["Giraud"],"firstnames":["Mathieu"],"suffixes":[]},{"propositions":[],"lastnames":["Groult"],"firstnames":["Richard"],"suffixes":[]},{"propositions":[],"lastnames":["Levé"],"firstnames":["Florence"],"suffixes":[]}],"editor":[{"propositions":[],"lastnames":["Mandel"],"firstnames":["Michael","I."],"suffixes":[]},{"propositions":[],"lastnames":["Devaney"],"firstnames":["Johanna"],"suffixes":[]},{"propositions":[],"lastnames":["Turnbull"],"firstnames":["Douglas"],"suffixes":[]},{"propositions":[],"lastnames":["Tzanetakis"],"firstnames":["George"],"suffixes":[]}],"year":"2016","note":"tex.date-added: 2019-03-29 16:18:17 +0100 tex.date-modified: 2019-03-29 16:18:17 +0100","keywords":"#nosource, ⛔ No DOI found","pages":"164–170","bibtex":"@inproceedings{guiomard-kagan_improving_2016,\n\ttitle = {Improving {Voice} {Separation} by {Better} {Connecting} {Contigs}},\n\turl = {https://wp.nyu.edu/ismir2016/wp-content/uploads/sites/2294/2016/07/129 \\_Paper.pdf},\n\tabstract = {Separating a polyphonic symbolic score into monophonic voices or streams helps to understand the music and may simplify further pattern matching. One of the best ways to compute this separation, as proposed by Chew and Wu in 2005 [2], is to first identify contigs that are portions of the music score with a constant number of voices, then to progressively connect these contigs. This raises two ques-tions: Which contigs should be connected first? And, how should these two contigs be connected? Here we propose to answer simultaneously these two questions by consid-ering a set of musical features that measures the quality of any connection. The coefficients weighting the features are optimized through a genetic algorithm. We benchmark the resulting connection policy on corpora containing fugues of the Well-Tempered Clavier by J. S. Bach as well as on string quartets, and we compare it against previously pro-posed policies [2, 9]. The contig connection is improved, particularly when one takes into account the whole content of voice fragments to assess the quality of their possible connection.},\n\tbooktitle = {Proceedings of the 17{\\textbackslash}textsuperscriptth {International} {Society} for {Music} {Information} {Retrieval} {Conference}, {ISMIR} 2016, {New} {York} {City}, {United} {States}, {August} 7-11, 2016},\n\tauthor = {Guiomard-Kagan, Nicolas and Giraud, Mathieu and Groult, Richard and Levé, Florence},\n\teditor = {Mandel, Michael I. and Devaney, Johanna and Turnbull, Douglas and Tzanetakis, George},\n\tyear = {2016},\n\tnote = {tex.date-added: 2019-03-29 16:18:17 +0100\ntex.date-modified: 2019-03-29 16:18:17 +0100},\n\tkeywords = {\\#nosource, ⛔ No DOI found},\n\tpages = {164--170},\n}\n\n","author_short":["Guiomard-Kagan, N.","Giraud, M.","Groult, R.","Levé, F."],"editor_short":["Mandel, M. I.","Devaney, J.","Turnbull, D.","Tzanetakis, G."],"key":"guiomard-kagan_improving_2016","id":"guiomard-kagan_improving_2016","bibbaseid":"guiomardkagan-giraud-groult-lev-improvingvoiceseparationbybetterconnectingcontigs-2016","role":"author","urls":{"Paper":"https://wp.nyu.edu/ismir2016/wp-content/uploads/sites/2294/2016/07/129 \\_Paper.pdf"},"keyword":["#nosource","⛔ No DOI found"],"metadata":{"authorlinks":{}},"html":""},"bibtype":"inproceedings","biburl":"https://bibbase.org/zotero/fsimonetta","dataSources":["pzyFFGWvxG2bs63zP"],"keywords":["#nosource","⛔ no doi found"],"search_terms":["improving","voice","separation","better","connecting","contigs","guiomard-kagan","giraud","groult","levé"],"title":"Improving Voice Separation by Better Connecting Contigs","year":2016}