Systematic Analysis of Music Representations from BERT

There have been numerous attempts to represent raw data as numerical vectors that effectively capture semantic and contextual information. However, in the field of symbolic music, previous works have attempted to validate their music embeddings by observing the performance improvement of various fin...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: Han, Sangjun, Ihm, Hyeongrae, Lim, Woohyung
Format: Artikel
Sprache:eng
Schlagworte:
Online-Zugang:Volltext bestellen
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
container_end_page
container_issue
container_start_page
container_title
container_volume
creator Han, Sangjun
Ihm, Hyeongrae
Lim, Woohyung
description There have been numerous attempts to represent raw data as numerical vectors that effectively capture semantic and contextual information. However, in the field of symbolic music, previous works have attempted to validate their music embeddings by observing the performance improvement of various fine-tuning tasks. In this work, we directly analyze embeddings from BERT and BERT with contrastive learning trained on bar-level MIDI, inspecting their musical information that can be obtained from MIDI events. We observe that the embeddings exhibit distinct characteristics of information depending on the contrastive objectives and the choice of layers. Our code is available at https://github.com/sjhan91/MusicBERT.
doi_str_mv 10.48550/arxiv.2306.04628
format Article
fullrecord <record><control><sourceid>arxiv_GOX</sourceid><recordid>TN_cdi_arxiv_primary_2306_04628</recordid><sourceformat>XML</sourceformat><sourcesystem>PC</sourcesystem><sourcerecordid>2306_04628</sourcerecordid><originalsourceid>FETCH-LOGICAL-a678-c4bd8f214c02abb08a487d99e9ce72191ac4c8156d4100226203c20e9f7e55013</originalsourceid><addsrcrecordid>eNotj81KxDAUhbNxIaMP4Mq8QOvNbZomy3EYf2BEGLsvt-kNFKbtkFSxb28dXR04Bw7fJ8SdglzbsoQHit_9V44FmBy0QXstzMeSZh5o7r3cjnRaUp_kFOTbZ1qbI58jJx7ndZ_GJEOcBvm4P9Y34irQKfHtf25E_bSvdy_Z4f35dbc9ZGQqm3nddjag0h6Q2hYsaVt1zrHzXKFyirz2VpWm0woA0SAUHoFdqHjFVcVG3P_dXsCbc-wHikvzK9BcBIof2zE_Qw</addsrcrecordid><sourcetype>Open Access Repository</sourcetype><iscdi>true</iscdi><recordtype>article</recordtype></control><display><type>article</type><title>Systematic Analysis of Music Representations from BERT</title><source>arXiv.org</source><creator>Han, Sangjun ; Ihm, Hyeongrae ; Lim, Woohyung</creator><creatorcontrib>Han, Sangjun ; Ihm, Hyeongrae ; Lim, Woohyung</creatorcontrib><description>There have been numerous attempts to represent raw data as numerical vectors that effectively capture semantic and contextual information. However, in the field of symbolic music, previous works have attempted to validate their music embeddings by observing the performance improvement of various fine-tuning tasks. In this work, we directly analyze embeddings from BERT and BERT with contrastive learning trained on bar-level MIDI, inspecting their musical information that can be obtained from MIDI events. We observe that the embeddings exhibit distinct characteristics of information depending on the contrastive objectives and the choice of layers. Our code is available at https://github.com/sjhan91/MusicBERT.</description><identifier>DOI: 10.48550/arxiv.2306.04628</identifier><language>eng</language><subject>Computer Science - Multimedia ; Computer Science - Sound</subject><creationdate>2023-06</creationdate><rights>http://creativecommons.org/licenses/by/4.0</rights><oa>free_for_read</oa><woscitedreferencessubscribed>false</woscitedreferencessubscribed></display><links><openurl>$$Topenurl_article</openurl><openurlfulltext>$$Topenurlfull_article</openurlfulltext><thumbnail>$$Tsyndetics_thumb_exl</thumbnail><link.rule.ids>228,230,780,885</link.rule.ids><linktorsrc>$$Uhttps://arxiv.org/abs/2306.04628$$EView_record_in_Cornell_University$$FView_record_in_$$GCornell_University$$Hfree_for_read</linktorsrc><backlink>$$Uhttps://doi.org/10.48550/arXiv.2306.04628$$DView paper in arXiv$$Hfree_for_read</backlink></links><search><creatorcontrib>Han, Sangjun</creatorcontrib><creatorcontrib>Ihm, Hyeongrae</creatorcontrib><creatorcontrib>Lim, Woohyung</creatorcontrib><title>Systematic Analysis of Music Representations from BERT</title><description>There have been numerous attempts to represent raw data as numerical vectors that effectively capture semantic and contextual information. However, in the field of symbolic music, previous works have attempted to validate their music embeddings by observing the performance improvement of various fine-tuning tasks. In this work, we directly analyze embeddings from BERT and BERT with contrastive learning trained on bar-level MIDI, inspecting their musical information that can be obtained from MIDI events. We observe that the embeddings exhibit distinct characteristics of information depending on the contrastive objectives and the choice of layers. Our code is available at https://github.com/sjhan91/MusicBERT.</description><subject>Computer Science - Multimedia</subject><subject>Computer Science - Sound</subject><fulltext>true</fulltext><rsrctype>article</rsrctype><creationdate>2023</creationdate><recordtype>article</recordtype><sourceid>GOX</sourceid><recordid>eNotj81KxDAUhbNxIaMP4Mq8QOvNbZomy3EYf2BEGLsvt-kNFKbtkFSxb28dXR04Bw7fJ8SdglzbsoQHit_9V44FmBy0QXstzMeSZh5o7r3cjnRaUp_kFOTbZ1qbI58jJx7ndZ_GJEOcBvm4P9Y34irQKfHtf25E_bSvdy_Z4f35dbc9ZGQqm3nddjag0h6Q2hYsaVt1zrHzXKFyirz2VpWm0woA0SAUHoFdqHjFVcVG3P_dXsCbc-wHikvzK9BcBIof2zE_Qw</recordid><startdate>20230606</startdate><enddate>20230606</enddate><creator>Han, Sangjun</creator><creator>Ihm, Hyeongrae</creator><creator>Lim, Woohyung</creator><scope>AKY</scope><scope>GOX</scope></search><sort><creationdate>20230606</creationdate><title>Systematic Analysis of Music Representations from BERT</title><author>Han, Sangjun ; Ihm, Hyeongrae ; Lim, Woohyung</author></sort><facets><frbrtype>5</frbrtype><frbrgroupid>cdi_FETCH-LOGICAL-a678-c4bd8f214c02abb08a487d99e9ce72191ac4c8156d4100226203c20e9f7e55013</frbrgroupid><rsrctype>articles</rsrctype><prefilter>articles</prefilter><language>eng</language><creationdate>2023</creationdate><topic>Computer Science - Multimedia</topic><topic>Computer Science - Sound</topic><toplevel>online_resources</toplevel><creatorcontrib>Han, Sangjun</creatorcontrib><creatorcontrib>Ihm, Hyeongrae</creatorcontrib><creatorcontrib>Lim, Woohyung</creatorcontrib><collection>arXiv Computer Science</collection><collection>arXiv.org</collection></facets><delivery><delcategory>Remote Search Resource</delcategory><fulltext>fulltext_linktorsrc</fulltext></delivery><addata><au>Han, Sangjun</au><au>Ihm, Hyeongrae</au><au>Lim, Woohyung</au><format>journal</format><genre>article</genre><ristype>JOUR</ristype><atitle>Systematic Analysis of Music Representations from BERT</atitle><date>2023-06-06</date><risdate>2023</risdate><abstract>There have been numerous attempts to represent raw data as numerical vectors that effectively capture semantic and contextual information. However, in the field of symbolic music, previous works have attempted to validate their music embeddings by observing the performance improvement of various fine-tuning tasks. In this work, we directly analyze embeddings from BERT and BERT with contrastive learning trained on bar-level MIDI, inspecting their musical information that can be obtained from MIDI events. We observe that the embeddings exhibit distinct characteristics of information depending on the contrastive objectives and the choice of layers. Our code is available at https://github.com/sjhan91/MusicBERT.</abstract><doi>10.48550/arxiv.2306.04628</doi><oa>free_for_read</oa></addata></record>
fulltext fulltext_linktorsrc
identifier DOI: 10.48550/arxiv.2306.04628
ispartof
issn
language eng
recordid cdi_arxiv_primary_2306_04628
source arXiv.org
subjects Computer Science - Multimedia
Computer Science - Sound
title Systematic Analysis of Music Representations from BERT
url https://sfx.bib-bvb.de/sfx_tum?ctx_ver=Z39.88-2004&ctx_enc=info:ofi/enc:UTF-8&ctx_tim=2024-12-25T22%3A07%3A15IST&url_ver=Z39.88-2004&url_ctx_fmt=infofi/fmt:kev:mtx:ctx&rfr_id=info:sid/primo.exlibrisgroup.com:primo3-Article-arxiv_GOX&rft_val_fmt=info:ofi/fmt:kev:mtx:journal&rft.genre=article&rft.atitle=Systematic%20Analysis%20of%20Music%20Representations%20from%20BERT&rft.au=Han,%20Sangjun&rft.date=2023-06-06&rft_id=info:doi/10.48550/arxiv.2306.04628&rft_dat=%3Carxiv_GOX%3E2306_04628%3C/arxiv_GOX%3E%3Curl%3E%3C/url%3E&disable_directlink=true&sfx.directlink=off&sfx.report_link=0&rft_id=info:oai/&rft_id=info:pmid/&rfr_iscdi=true