{"id":209766,"created":"2025-01-19T01:11:03.266735+00:00","metadata":{"_oai":{"id":"oai:ipsj.ixsq.nii.ac.jp:00209766","sets":["1164:5159:10515:10530"]},"path":["10530"],"owner":"44499","recid":"209766","title":["NMF基底間の識別性に関する定量的尺度"],"pubdate":{"attribute_name":"公開日","attribute_value":"2021-02-24"},"_buckets":{"deposit":"b95812d0-57d7-4994-bdac-3e18aa2d488a"},"_deposit":{"id":"209766","pid":{"type":"depid","value":"209766","revision_id":0},"owners":[44499],"status":"published","created_by":44499},"item_title":"NMF基底間の識別性に関する定量的尺度","author_link":["529663","529661","529660","529662","529665","529664"],"item_titles":{"attribute_name":"タイトル","attribute_value_mlt":[{"subitem_title":"NMF基底間の識別性に関する定量的尺度"},{"subitem_title":"A quantitative measure of discriminability between NMF dictionaries","subitem_title_language":"en"}]},"item_keyword":{"attribute_name":"キーワード","attribute_value_mlt":[{"subitem_subject":"EA1","subitem_subject_scheme":"Other"}]},"item_type_id":"4","publish_date":"2021-02-24","item_4_text_3":{"attribute_name":"著者所属","attribute_value_mlt":[{"subitem_text_value":"東京大学"},{"subitem_text_value":"東京大学"},{"subitem_text_value":"東京大学"}]},"item_4_text_4":{"attribute_name":"著者所属(英)","attribute_value_mlt":[{"subitem_text_value":"The University of Tokyo","subitem_text_language":"en"},{"subitem_text_value":"The University of Tokyo","subitem_text_language":"en"},{"subitem_text_value":"The 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/209766/files/IPSJ-SLP21136028.pdf","label":"IPSJ-SLP21136028.pdf"},"format":"application/pdf","billing":["billing_file"],"filename":"IPSJ-SLP21136028.pdf","filesize":[{"value":"1.7 MB"}],"mimetype":"application/pdf","priceinfo":[{"tax":["include_tax"],"price":"0","billingrole":"22"},{"tax":["include_tax"],"price":"0","billingrole":"44"}],"accessrole":"open_login","version_id":"9b7293c3-5622-4482-9f72-a41962381342","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":"紺野, 瑛介"}],"nameIdentifiers":[{}]},{"creatorNames":[{"creatorName":"齋藤, 大輔"}],"nameIdentifiers":[{}]},{"creatorNames":[{"creatorName":"峯松, 信明"}],"nameIdentifiers":[{}]}]},"item_4_creator_6":{"attribute_name":"著者名(英)","attribute_type":"creator","attribute_value_mlt":[{"creatorNames":[{"creatorName":"Eisuke, Konno","creatorNameLang":"en"}],"nameIdentifiers":[{}]},{"creatorNames":[{"creatorName":"Daisuke, Saito","creatorNameLang":"en"}],"nameIdentifiers":[{}]},{"creatorNames":[{"creatorName":"Nobuaki, Minematsu","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":"モノラル音源分離の代表的な手法として教師あり非負値行列因子分解（Nonnegative Matrix Factorization; NMF）がある．教師ありNMFでは各音源の基底を事前に学習しておくことで分離を実現する．分離対象が音声同士のような類似した音源群の場合に有効な基底学習法として識別的 NMF  が提案されているが，これが指す「識別性」とは何かということについて定量的な議論は今までなされてこなかった．そこで本研究では，ある基底学習法 により得られる各音源基底間の「識別性」を定量的に測るオーバーラップ尺度を提案し，これによる基底学習法間 の比較の基準として，各音源の真の基底の間で測ったこの尺度を用いることの妥当性を論じる．真の基底は一般に 未知だが，最小体積NMF  により近似的に得ることができる．オラクルの学習データを用いたモノラル音声分離実験の結果，識別的NMFはこのオーバーラップ尺度を小さくするという意味で「識別的」な基底を学習し，それにより分離性能を高めていると考えられることがわかった．しかし，基準である最小体積NMFの方が，オーバーラップ尺度がより小さく，分離性能はより高いという結果となり，実は基底学習法としてより識別的で優れているという可 能性が示唆された．","subitem_description_type":"Other"}]},"item_4_biblio_info_10":{"attribute_name":"書誌情報","attribute_value_mlt":[{"bibliographicPageEnd":"6","bibliographic_titles":[{"bibliographic_title":"研究報告音声言語情報処理（SLP）"}],"bibliographicPageStart":"1","bibliographicIssueDates":{"bibliographicIssueDate":"2021-02-24","bibliographicIssueDateType":"Issued"},"bibliographicIssueNumber":"28","bibliographicVolumeNumber":"2021-SLP-136"}]},"relation_version_is_last":true,"weko_creator_id":"44499"},"updated":"2025-01-19T18:23:53.875714+00:00","links":{}}