{"links":{},"metadata":{"_oai":{"id":"oai:ipsj.ixsq.nii.ac.jp:00056046","sets":["1164:5064:5095:5098"]},"path":["5098"],"owner":"1","recid":"56046","title":["ハーモニック・クラスタリングによる多重音信号音高抽出における音源数とオクターブ位置推定"],"pubdate":{"attribute_name":"公開日","attribute_value":"2003-08-04"},"_buckets":{"deposit":"0e56ac64-9931-45a6-bd07-f665b284f296"},"_deposit":{"id":"56046","pid":{"type":"depid","value":"56046","revision_id":0},"owners":[1],"status":"published","created_by":1},"item_title":"ハーモニック・クラスタリングによる多重音信号音高抽出における音源数とオクターブ位置推定","author_link":["0","0"],"item_titles":{"attribute_name":"タイトル","attribute_value_mlt":[{"subitem_title":"ハーモニック・クラスタリングによる多重音信号音高抽出における音源数とオクターブ位置推定"},{"subitem_title":"Estimation of Number of Sound Sources and Octave Position in Multi - Pitch Extraction Using Harmonic Clustering","subitem_title_language":"en"}]},"item_type_id":"4","publish_date":"2003-08-04","item_4_text_3":{"attribute_name":"著者所属","attribute_value_mlt":[{"subitem_text_value":"東京大学大学院情報理工学系研究科"},{"subitem_text_value":"東京大学大学院情報理工学系研究科"},{"subitem_text_value":"東京大学大学院情報理工学系研究科"}]},"item_4_text_4":{"attribute_name":"著者所属(英)","attribute_value_mlt":[{"subitem_text_value":"Graduate School of Information Science and Technology, University of Tokyo","subitem_text_language":"en"},{"subitem_text_value":"Graduate School of Information Science and Technology, University of Tokyo","subitem_text_language":"en"},{"subitem_text_value":"Graduate School of Information Science and Technology, University of Tokyo","subitem_text_language":"en"}]},"item_language":{"attribute_name":"言語","attribute_value_mlt":[{"subitem_language":"jpn"}]},"item_publisher":{"attribute_name":"出版者","attribute_value_mlt":[{"subitem_publisher":"情報処理学会","subitem_publisher_language":"ja"}]},"publish_status":"0","weko_shared_id":-1,"item_file_price":{"attribute_name":"Billing file","attribute_type":"file","attribute_value_mlt":[{"url":{"url":"https://ipsj.ixsq.nii.ac.jp/record/56046/files/IPSJ-MUS03051006.pdf"},"date":[{"dateType":"Available","dateValue":"2005-08-04"}],"format":"application/pdf","billing":["billing_file"],"filename":"IPSJ-MUS03051006.pdf","filesize":[{"value":"293.9 kB"}],"mimetype":"application/pdf","priceinfo":[{"tax":["include_tax"],"price":"660","billingrole":"5"},{"tax":["include_tax"],"price":"330","billingrole":"6"},{"tax":["include_tax"],"price":"0","billingrole":"21"},{"tax":["include_tax"],"price":"0","billingrole":"44"}],"accessrole":"open_date","version_id":"a9931c92-495b-4448-b8ae-383bd8b1a858","displaytype":"detail","licensetype":"license_note","license_note":"Copyright (c) 2003 by the Information Processing Society of Japan"}]},"item_4_creator_5":{"attribute_name":"著者名","attribute_type":"creator","attribute_value_mlt":[{"creatorNames":[{"creatorName":"亀岡, 弘和"},{"creatorName":"西本, 卓也"},{"creatorName":"嵯峨山, 茂樹"}],"nameIdentifiers":[{}]}]},"item_4_creator_6":{"attribute_name":"著者名(英)","attribute_type":"creator","attribute_value_mlt":[{"creatorNames":[{"creatorName":"Hirokazu, Kameoka","creatorNameLang":"en"},{"creatorName":"Takuya, Nishimoto","creatorNameLang":"en"},{"creatorName":"Shigeki, Sagayama","creatorNameLang":"en"}],"nameIdentifiers":[{}]}]},"item_4_source_id_9":{"attribute_name":"書誌レコードID","attribute_value_mlt":[{"subitem_source_identifier":"AN10438388","subitem_source_identifier_type":"NCID"}]},"item_4_textarea_12":{"attribute_name":"Notice","attribute_value_mlt":[{"subitem_textarea_value":"SIG Technical Reports are nonrefereed and hence may later appear in any journals, conferences, symposia, etc."}]},"item_resource_type":{"attribute_name":"資源タイプ","attribute_value_mlt":[{"resourceuri":"http://purl.org/coar/resource_type/c_18gh","resourcetype":"technical report"}]},"item_4_description_7":{"attribute_name":"論文抄録","attribute_value_mlt":[{"subitem_description":"本報告は，多重音のモノラル音響信号の基本周波数抽出における音源数推定およびオクターブ位置推定を扱う．ある周波数におけるパワースペクトル密度は，ある一定の大きさの微小エネルギーを多数個加算したものと考え，その微小エネルギーに対し周波数軸上でのクラスタリングを適用する．まず単一音の場合，単純なクラスタリングではなく，倍音構造としての拘束をもつ複数のクラスタから成るクラスタ群の決定により基本周波数を抽出できる．このアプローチを我々はハーモニック・クラスタリングと呼ぶ．次に，微小エネルギーのクラスタ帰属度を正規分布に基づく確率的な尺度で扱い，同様な拘束条件をもつ複数のクラスタ群を同時に決定する問題をEMアルゴリズムにより解くことで，異なる音源同士で重なり合う倍音スペクトル成分を確率的に分解し，複数の基本周波数を抽出できる．我々はこれまでに，ハーモニック・クラスタリングを適用したピッチトラッキングのための実験系を構成し，収束性能の確認を行った．本報告では，これまで扱わなかった音源数およびオクターブ位置推定の問題を対象にし，クラスタリングの過程においてAICに基づいてこれらを推定する方法を検討する．これを取り入れた基本周波数抽出アルゴリズムの動作確認を，実演奏を録音したモノラル音響信号を対象とした実験により行う．","subitem_description_type":"Other"}]},"item_4_description_8":{"attribute_name":"論文抄録(英)","attribute_value_mlt":[{"subitem_description":"This report describes a method for estimating the number of sound sources and octave position in the process of extracting multiple fundamental frequencies of monaural audio signals. Considering that power spectral densities are sum total of a large number of micro-energies, clustering can be applied to the micro-energies in the frequency domain. We call this approach “Harmonic Clustering”. In case of single pitch, fundamentalfrequency can be estimated by determining a group of clusters which consist of multiple clusters having a constraint for constructing a harmonic structure. It is further extended to the case of multiple pitches by using a probablistic measure based on Gaussian distributions for resolving the micro-energies to each cluster so as to separate overlapped harmonic spectral components and by giving multiple cluster groups each having the same constraint.  Determining problem of the cluster groups can be formulated in a way similar to the EM algorithm. We have constructed a pitch tracking setup based on the Harmonic Clustering and have evaluated its convergence up to now. In this report, we focus on the problem of the number of sound sources and the octave position and propose methods for estimating them using AIC calculated during the process of the clustering. We experimented on extracting the fundamental frequencies of monaural musical audio signals by integrating these estimation methods stated above.","subitem_description_type":"Other"}]},"item_4_biblio_info_10":{"attribute_name":"書誌情報","attribute_value_mlt":[{"bibliographicPageEnd":"34","bibliographic_titles":[{"bibliographic_title":"情報処理学会研究報告音楽情報科学（MUS）"}],"bibliographicPageStart":"29","bibliographicIssueDates":{"bibliographicIssueDate":"2003-08-04","bibliographicIssueDateType":"Issued"},"bibliographicIssueNumber":"82(2003-MUS-051)","bibliographicVolumeNumber":"2003"}]},"relation_version_is_last":true,"weko_creator_id":"1"},"created":"2025-01-18T23:19:33.502292+00:00","updated":"2025-01-22T05:09:30.769290+00:00","id":56046}