{"metadata":{"_oai":{"id":"oai:ipsj.ixsq.nii.ac.jp:00241585","sets":["1164:4179:11560:11869"]},"path":["11869"],"owner":"44499","recid":"241585","title":["役割情報付与に基づく重要文強調音声合成―ピッチ予測の重みづけによる強調度合の制御"],"pubdate":{"attribute_name":"公開日","attribute_value":"2024-12-05"},"_buckets":{"deposit":"691a3827-855b-4a94-b687-88c507e7b67a"},"_deposit":{"id":"241585","pid":{"type":"depid","value":"241585","revision_id":0},"owners":[44499],"status":"published","created_by":44499},"item_title":"役割情報付与に基づく重要文強調音声合成―ピッチ予測の重みづけによる強調度合の制御","author_link":["665359","665362","665361","665363","665360","665364"],"item_titles":{"attribute_name":"タイトル","attribute_value_mlt":[{"subitem_title":"役割情報付与に基づく重要文強調音声合成―ピッチ予測の重みづけによる強調度合の制御"},{"subitem_title":"Role-driven speech synthesis emphasizing nucleus sentences―Emphasis control by weighting pitch prediction","subitem_title_language":"en"}]},"item_keyword":{"attribute_name":"キーワード","attribute_value_mlt":[{"subitem_subject":"ポスターセッション","subitem_subject_scheme":"Other"}]},"item_type_id":"4","publish_date":"2024-12-05","item_4_text_3":{"attribute_name":"著者所属","attribute_value_mlt":[{"subitem_text_value":"早稲田大学"},{"subitem_text_value":"千葉工業大学"},{"subitem_text_value":"早稲田大学"}]},"item_4_text_4":{"attribute_name":"著者所属(英)","attribute_value_mlt":[{"subitem_text_value":"Waseda University","subitem_text_language":"en"},{"subitem_text_value":"Chiba Institute of Technology","subitem_text_language":"en"},{"subitem_text_value":"Waseda University","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/241585/files/IPSJ-NL24262010.pdf","label":"IPSJ-NL24262010.pdf"},"date":[{"dateType":"Available","dateValue":"2026-12-05"}],"format":"application/pdf","billing":["billing_file"],"filename":"IPSJ-NL24262010.pdf","filesize":[{"value":"422.2 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":"23"},{"tax":["include_tax"],"price":"0","billingrole":"44"}],"accessrole":"open_date","version_id":"8f530c01-ba6b-4613-aed0-e742924ed664","displaytype":"detail","licensetype":"license_note","license_note":"Copyright (c) 2024 by the Information Processing Society of Japan"}]},"item_4_creator_5":{"attribute_name":"著者名","attribute_type":"creator","attribute_value_mlt":[{"creatorNames":[{"creatorName":"平田, 瑞葵"}],"nameIdentifiers":[{}]},{"creatorNames":[{"creatorName":"藤江, 真也"}],"nameIdentifiers":[{}]},{"creatorNames":[{"creatorName":"小林, 哲則"}],"nameIdentifiers":[{}]}]},"item_4_creator_6":{"attribute_name":"著者名(英)","attribute_type":"creator","attribute_value_mlt":[{"creatorNames":[{"creatorName":"Mizuki, Hirata","creatorNameLang":"en"}],"nameIdentifiers":[{}]},{"creatorNames":[{"creatorName":"Shinya, Fujie","creatorNameLang":"en"}],"nameIdentifiers":[{}]},{"creatorNames":[{"creatorName":"Tetsunori, Kobayashi","creatorNameLang":"en"}],"nameIdentifiers":[{}]}]},"item_4_source_id_9":{"attribute_name":"書誌レコードID","attribute_value_mlt":[{"subitem_source_identifier":"AN10115061","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":"2188-8779","subitem_source_identifier_type":"ISSN"}]},"item_4_description_7":{"attribute_name":"論文抄録","attribute_value_mlt":[{"subitem_description":"前置き，核，補足など，各文章の段落内での役割を補助情報として与えることで，段落文章をメリハリ良く発話できる音声合成方式を実現する．これまで，韻律パラメタ生成と音声合成を二段階で行う Merlin，End-to-End 方式の VITS を対象に検討してきたが，前者には音質の問題が，後者には役割に応じたメリハリを十分表現できないという問題があった．ここでは音質が良いとされる End-to-End 方式の中で，韻律制御部を独立して持つことに特徴を持つ JETS を対象とした．JETS では，韻律の違いが韻律制御部に集約して学習されることから，韻律変化の特徴を捉え易くなることが期待できる．実験の結果，JETS は他手法に比べ役割に応じた韻律変化を良く表現できること，学習時に韻律制御部の損失の重みを大きくすることで，核の部分を更に強調できることが確認できた．","subitem_description_type":"Other"}]},"item_4_biblio_info_10":{"attribute_name":"書誌情報","attribute_value_mlt":[{"bibliographicPageEnd":"6","bibliographic_titles":[{"bibliographic_title":"研究報告自然言語処理（NL）"}],"bibliographicPageStart":"1","bibliographicIssueDates":{"bibliographicIssueDate":"2024-12-05","bibliographicIssueDateType":"Issued"},"bibliographicIssueNumber":"10","bibliographicVolumeNumber":"2024-NL-262"}]},"relation_version_is_last":true,"weko_creator_id":"44499"},"id":241585,"updated":"2025-01-19T07:36:47.390201+00:00","links":{},"created":"2025-01-19T01:46:16.617377+00:00"}