{"created":"2025-01-19T00:51:10.986019+00:00","metadata":{"_oai":{"id":"oai:ipsj.ixsq.nii.ac.jp:00183679","sets":["1164:5159:9063:9265"]},"path":["9265"],"owner":"11","recid":"183679","title":["雑音環境下音声を用いた音声合成のための雑音生成モデルの敵対的学習"],"pubdate":{"attribute_name":"公開日","attribute_value":"2017-10-06"},"_buckets":{"deposit":"d63f768f-dca1-4d12-81f3-6b50f4d6ddfb"},"_deposit":{"id":"183679","pid":{"type":"depid","value":"183679","revision_id":0},"owners":[11],"status":"published","created_by":11},"item_title":"雑音環境下音声を用いた音声合成のための雑音生成モデルの敵対的学習","author_link":["403918","403919","403910","403911","403921","403914","403912","403913","403920","403915","403917","403916"],"item_titles":{"attribute_name":"タイトル","attribute_value_mlt":[{"subitem_title":"雑音環境下音声を用いた音声合成のための雑音生成モデルの敵対的学習"},{"subitem_title":"Generative adversarial training of the noise generation model for speech synthesis using speech in noise","subitem_title_language":"en"}]},"item_type_id":"4","publish_date":"2017-10-06","item_4_text_3":{"attribute_name":"著者所属","attribute_value_mlt":[{"subitem_text_value":"徳山工業高等専門学校／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"},{"subitem_text_value":"東京大学大学院情報理工学系研究科"},{"subitem_text_value":"徳山工業高等専門学校"},{"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":"National Institute of Technology, Tokuyama College / 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"},{"subitem_text_value":"The University of Tokyo","subitem_text_language":"en"},{"subitem_text_value":"National Institute of Technology, Tokuyama College","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/183679/files/IPSJ-SLP17118001.pdf","label":"IPSJ-SLP17118001.pdf"},"date":[{"dateType":"Available","dateValue":"2019-10-06"}],"format":"application/pdf","billing":["billing_file"],"filename":"IPSJ-SLP17118001.pdf","filesize":[{"value":"1.3 MB"}],"mimetype":"application/pdf","priceinfo":[{"tax":["include_tax"],"price":"660","billingrole":"5"},{"tax":["include_tax"],"price":"330","billingrole":"6"},{"tax":["include_tax"],"price":"0","billingrole":"22"},{"tax":["include_tax"],"price":"0","billingrole":"44"}],"accessrole":"open_date","version_id":"53f5556d-4a67-4e26-bd4c-f24145bb8015","displaytype":"detail","licensetype":"license_note","license_note":"Copyright (c) 2017 by the Information Processing Society of Japan"}]},"item_4_creator_5":{"attribute_name":"著者名","attribute_type":"creator","attribute_value_mlt":[{"creatorNames":[{"creatorName":"宇根, 昌和"}],"nameIdentifiers":[{}]},{"creatorNames":[{"creatorName":"齋藤, 佑樹"}],"nameIdentifiers":[{}]},{"creatorNames":[{"creatorName":"高道, 慎之介"}],"nameIdentifiers":[{}]},{"creatorNames":[{"creatorName":"北村, 大地"}],"nameIdentifiers":[{}]},{"creatorNames":[{"creatorName":"宮崎, 亮一"}],"nameIdentifiers":[{}]},{"creatorNames":[{"creatorName":"猿渡, 洋"}],"nameIdentifiers":[{}]}]},"item_4_creator_6":{"attribute_name":"著者名(英)","attribute_type":"creator","attribute_value_mlt":[{"creatorNames":[{"creatorName":"Masakazu, Une","creatorNameLang":"en"}],"nameIdentifiers":[{}]},{"creatorNames":[{"creatorName":"Yuki, Saito","creatorNameLang":"en"}],"nameIdentifiers":[{}]},{"creatorNames":[{"creatorName":"Shinnnosuke, Takamichi","creatorNameLang":"en"}],"nameIdentifiers":[{}]},{"creatorNames":[{"creatorName":"Daichi, Kitamura","creatorNameLang":"en"}],"nameIdentifiers":[{}]},{"creatorNames":[{"creatorName":"Ryoichi, Miyazaki","creatorNameLang":"en"}],"nameIdentifiers":[{}]},{"creatorNames":[{"creatorName":"Hiroshi, Saruwatari","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":"高品質な統計的パラメトリック音声合成システムの構築には，スタジオ等の理想的な環境で収録された音声データの利用が不可欠であるため，現存する膨大な音声データのうち，音声合成の学習に利用可能なものは非常に限定される．本稿では，雑音環境下音声から高品質な音声合成を構築する方法を提案する．従来，そのような音声を学習データとして用いる場合，spectral subtraction 等の雑音抑圧処理を施した後に，通常の音声合成の学習を行う．しかしながら，雑音スペクトルの生成分布をパラメトリックに定義する雑音抑圧法は処理後の音声を歪ませ，さらに，その歪みは音声合成の学習時に増幅されて合成音声品質を悪化させる．そこで本稿では，敵対的学習アルゴリズムにより学習される雑音生成モデルを用いた，音声合成の学習法を提案する．雑音生成モデルは，観測雑音スペクトルの統計量を持つように学習され，雑音スペクトルを確率的に生成する．テキストから音声スペクトルを生成する音声合成モデルは，生成雑音を加算した後のスペクトルが雑音環境下音声のスペクトルに一致するように学習される．提案法は，雑音スペクトルの生成分布を柔軟にモデル化でき，さらに，雑音加算過程を考慮して音声合成モデルを学習するため，従来法において生じる品質低下を低減できる．実験的評価では，いくつかの雑音抑圧強度と SN 比において合成音声を作成し，提案法の知覚的音質が従来法を上回ることを示す．","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":"2017-10-06","bibliographicIssueDateType":"Issued"},"bibliographicIssueNumber":"1","bibliographicVolumeNumber":"2017-SLP-118"}]},"relation_version_is_last":true,"weko_creator_id":"11"},"links":{},"id":183679,"updated":"2025-01-20T03:34:18.610294+00:00"}