{"updated":"2025-01-21T18:34:45.619818+00:00","metadata":{"_oai":{"id":"oai:ipsj.ixsq.nii.ac.jp:00083402","sets":["1164:5064:6681:6843"]},"path":["6843"],"owner":"11","recid":"83402","title":["VocaListenerによる学習データ生成を利用した多対多固有声変換に基づく歌声声質変換"],"pubdate":{"attribute_name":"公開日","attribute_value":"2012-08-02"},"_buckets":{"deposit":"4f014bc4-5150-4eca-a12d-d8a3a1f2095f"},"_deposit":{"id":"83402","pid":{"type":"depid","value":"83402","revision_id":0},"owners":[11],"status":"published","created_by":11},"item_title":"VocaListenerによる学習データ生成を利用した多対多固有声変換に基づく歌声声質変換","author_link":["0"],"item_titles":{"attribute_name":"タイトル","attribute_value_mlt":[{"subitem_title":"VocaListenerによる学習データ生成を利用した多対多固有声変換に基づく歌声声質変換"}]},"item_keyword":{"attribute_name":"キーワード","attribute_value_mlt":[{"subitem_subject":"歌声情報処理:信号処理","subitem_subject_scheme":"Other"}]},"item_type_id":"4","publish_date":"2012-08-02","item_4_text_3":{"attribute_name":"著者所属","attribute_value_mlt":[{"subitem_text_value":"奈良先端科学技術大学院大学"},{"subitem_text_value":"奈良先端科学技術大学院大学"},{"subitem_text_value":"産業技術総合研究所"},{"subitem_text_value":"産業技術総合研究所"},{"subitem_text_value":"奈良先端科学技術大学院大学"}]},"item_4_text_4":{"attribute_name":"著者所属(英)","attribute_value_mlt":[{"subitem_text_value":"Graduate School of Information Science, Nara Institute of Science and Technology (NAIST)","subitem_text_language":"en"},{"subitem_text_value":"Graduate School of Information Science, Nara Institute of Science and Technology (NAIST)","subitem_text_language":"en"},{"subitem_text_value":"National Institute of Advanced Industrial Science and Technology (AIST)","subitem_text_language":"en"},{"subitem_text_value":"National Institute of Advanced Industrial Science and Technology (AIST)","subitem_text_language":"en"},{"subitem_text_value":"Graduate School of Information Science, Nara Institute of Science and Technology (NAIST)","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/83402/files/IPSJ-MUS12096005.pdf"},"date":[{"dateType":"Available","dateValue":"2014-08-02"}],"format":"application/pdf","billing":["billing_file"],"filename":"IPSJ-MUS12096005.pdf","filesize":[{"value":"413.8 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":"58b60a22-6d74-431b-a8e7-931352df20fe","displaytype":"detail","licensetype":"license_note","license_note":"Copyright (c) 2012 by the Information Processing Society of Japan"}]},"item_4_creator_5":{"attribute_name":"著者名","attribute_type":"creator","attribute_value_mlt":[{"creatorNames":[{"creatorName":"土井啓成"},{"creatorName":"戸田智基"},{"creatorName":"中野倫靖"},{"creatorName":"後藤真孝"},{"creatorName":"中村哲"}],"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":"歌声の声質には,歌手の個人性が反映されており,他者の声質に自在に切り替えて歌うことは難しい.そこで我々は,歌声の声質を他者の歌声の声質へと自動変換することで,任意の声質での歌唱を実現する手法を提案し,歌唱という音楽表現の可能性を広げることを目指す.従来,統計的声質変換に基づく歌声声質変換が実現されていたが,提案手法では様々な声質に少ない負担で変換可能にするため,多対多固有声変換を導入する.これにより変換時に数秒程度の少量の無伴奏歌声さえあれば,任意の歌手の歌声から別の任意の歌手の歌声への声質変換が実現できる.しかし,その声質変換モデルの事前学習データとして,ある参照歌手の歌声と多くの事前収録目標歌手の歌声とのペアから構成されるパラレルデータセットが必要で,その歌声収録は困難であった.そこで提案手法では,歌唱表現を模倣できる歌声合成システム VocaListener を用いて目標歌手の歌声から参照歌手の歌声を生成することで,その学習データ構築を容易にする.実験結果から提案手法の有効性を確認した.","subitem_description_type":"Other"}]},"item_4_biblio_info_10":{"attribute_name":"書誌情報","attribute_value_mlt":[{"bibliographicPageEnd":"9","bibliographic_titles":[{"bibliographic_title":"研究報告音楽情報科学(MUS)"}],"bibliographicPageStart":"1","bibliographicIssueDates":{"bibliographicIssueDate":"2012-08-02","bibliographicIssueDateType":"Issued"},"bibliographicIssueNumber":"5","bibliographicVolumeNumber":"2012-MUS-96"}]},"relation_version_is_last":true,"weko_creator_id":"11"},"created":"2025-01-18T23:36:54.195917+00:00","id":83402,"links":{}}