{"metadata":{"_oai":{"id":"oai:ipsj.ixsq.nii.ac.jp:00214110","sets":["1164:5159:10515:10760"]},"path":["10760"],"owner":"44499","recid":"214110","title":["CQCCを用いた正則化NMFによる音源分離"],"pubdate":{"attribute_name":"公開日","attribute_value":"2021-11-24"},"_buckets":{"deposit":"2c7955e7-70b8-4dde-9d5b-5211a3d1f1ed"},"_deposit":{"id":"214110","pid":{"type":"depid","value":"214110","revision_id":0},"owners":[44499],"status":"published","created_by":44499},"item_title":"CQCCを用いた正則化NMFによる音源分離","author_link":["548686","548685","548688","548687"],"item_titles":{"attribute_name":"タイトル","attribute_value_mlt":[{"subitem_title":"CQCCを用いた正則化NMFによる音源分離"},{"subitem_title":"Music Separation by Regularized NMF using CQCC","subitem_title_language":"en"}]},"item_keyword":{"attribute_name":"キーワード","attribute_value_mlt":[{"subitem_subject":"音声・音響特徴","subitem_subject_scheme":"Other"}]},"item_type_id":"4","publish_date":"2021-11-24","item_4_text_3":{"attribute_name":"著者所属","attribute_value_mlt":[{"subitem_text_value":"山梨大学"},{"subitem_text_value":"山梨大学"}]},"item_4_text_4":{"attribute_name":"著者所属(英)","attribute_value_mlt":[{"subitem_text_value":"University of Yamanashi","subitem_text_language":"en"},{"subitem_text_value":"University of Yamanashi","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/214110/files/IPSJ-SLP21139016.pdf","label":"IPSJ-SLP21139016.pdf"},"format":"application/pdf","billing":["billing_file"],"filename":"IPSJ-SLP21139016.pdf","filesize":[{"value":"1.2 MB"}],"mimetype":"application/pdf","priceinfo":[{"tax":["include_tax"],"price":"0","billingrole":"22"},{"tax":["include_tax"],"price":"0","billingrole":"44"}],"accessrole":"open_login","version_id":"b832f1d1-2231-4a34-97f4-bf3856044584","displaytype":"detail","licensetype":"license_note","license_note":"Copyright (c) 2021 by the Institute of Electronics, Information and Communication Engineers This SIG report is only available to those in membership of the SIG."}]},"item_4_creator_5":{"attribute_name":"著者名","attribute_type":"creator","attribute_value_mlt":[{"creatorNames":[{"creatorName":"宮島, 浩平"}],"nameIdentifiers":[{}]},{"creatorNames":[{"creatorName":"大木, 真"}],"nameIdentifiers":[{}]}]},"item_4_creator_6":{"attribute_name":"著者名(英)","attribute_type":"creator","attribute_value_mlt":[{"creatorNames":[{"creatorName":"Kohei, Miyajima","creatorNameLang":"en"}],"nameIdentifiers":[{}]},{"creatorNames":[{"creatorName":"Makoto, Ohki","creatorNameLang":"en"}],"nameIdentifiers":[{}]}]},"item_4_source_id_9":{"attribute_name":"書誌レコードID","attribute_value_mlt":[{"subitem_source_identifier":"AN10442647","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_source_id_11":{"attribute_name":"ISSN","attribute_value_mlt":[{"subitem_source_identifier":"2188-8663","subitem_source_identifier_type":"ISSN"}]},"item_4_description_7":{"attribute_name":"論文抄録","attribute_value_mlt":[{"subitem_description":"本研究では，個別に録音された複数の音源から成る楽曲を対象とし，音響特徴量を用いた教師ありの音源分離手法を提案する．音源分離手法の 1 つである非負値行列因子分解 (Non-negative Matrix Factorization: NMF) の拡張として，音響特徴量を用いて個々の楽器のスペクトル包絡が学習データのスペクトル包絡の分布に沿うような制約を加えたケプストラム距離正則化 (Cepstral Distance Regularization: CDR) が提案されている．音響特徴量として一般に用いられるメル周波数ケプストラム係数 (Mel-Frequency Cepstral Coefficient: MFCC) は，人間の音高知覚に基づいたケプストラム分析の 1 つであり，音声認識等に利用されている．MFCC の算出に用いられる短時間フーリエ変換 (Short-Time Fourier Transform: STFT) は，周波数分解能と時間分解能がトレードオフの関係にあるため，広い周波数帯域で構成されている楽曲においては，低周波数の分解能が十分でないという問題がある．したがって，提案手法では，どの周波数でも分解能が一定となるように窓幅を変更しながら時間周波数解析を行う定 Q 変換 (Constant-Q Transform: CQT) をケプストラム分析によって音響特徴量化した，定 Q ケプストラム係数 (Constant-Q Cepstral Coefficient: CQCC) を用いて音源分離を行う．","subitem_description_type":"Other"}]},"item_4_biblio_info_10":{"attribute_name":"書誌情報","attribute_value_mlt":[{"bibliographicPageEnd":"5","bibliographic_titles":[{"bibliographic_title":"研究報告音声言語情報処理（SLP）"}],"bibliographicPageStart":"1","bibliographicIssueDates":{"bibliographicIssueDate":"2021-11-24","bibliographicIssueDateType":"Issued"},"bibliographicIssueNumber":"16","bibliographicVolumeNumber":"2021-SLP-139"}]},"relation_version_is_last":true,"weko_creator_id":"44499"},"id":214110,"updated":"2025-01-19T16:53:53.662367+00:00","links":{},"created":"2025-01-19T01:14:56.611974+00:00"}