{"links":{},"id":228438,"metadata":{"_oai":{"id":"oai:ipsj.ixsq.nii.ac.jp:00228438","sets":["1164:5159:11151:11374"]},"path":["11374"],"owner":"44499","recid":"228438","title":["言語表現による喉頭摘出者のための音声強調システム"],"pubdate":{"attribute_name":"公開日","attribute_value":"2023-10-07"},"_buckets":{"deposit":"808ae6f9-d099-4b78-914c-61cf6c7c2835"},"_deposit":{"id":"228438","pid":{"type":"depid","value":"228438","revision_id":0},"owners":[44499],"status":"published","created_by":44499},"item_title":"言語表現による喉頭摘出者のための音声強調システム","author_link":["609760","609770","609761","609766","609768","609769","609765","609763","609764","609767","609762"],"item_titles":{"attribute_name":"タイトル","attribute_value_mlt":[{"subitem_title":"言語表現による喉頭摘出者のための音声強調システム"},{"subitem_title":"Electrolaryngeal Speech Enhancement Through Strong Linguistic Encoding Methods","subitem_title_language":"en"}]},"item_keyword":{"attribute_name":"キーワード","attribute_value_mlt":[{"subitem_subject":"音声","subitem_subject_scheme":"Other"}]},"item_type_id":"4","publish_date":"2023-10-07","item_4_text_3":{"attribute_name":"著者所属","attribute_value_mlt":[{"subitem_text_value":"名古屋大学情報学研究科知能システム学専攻"},{"subitem_text_value":"名古屋大学情報学研究科知能システム学専攻"},{"subitem_text_value":"名古屋大学情報学研究科知能システム学専攻"},{"subitem_text_value":"名古屋大学情報学研究科知能システム学専攻／株式会社TARVO"},{"subitem_text_value":"名古屋大学情報基盤センター"}]},"item_4_text_4":{"attribute_name":"著者所属(英)","attribute_value_mlt":[{"subitem_text_value":"Graduate School of Informatics, Nagoya University","subitem_text_language":"en"},{"subitem_text_value":"Graduate School of Informatics, Nagoya University","subitem_text_language":"en"},{"subitem_text_value":"Graduate School of Informatics, Nagoya University","subitem_text_language":"en"},{"subitem_text_value":"Graduate School of Informatics, Nagoya University / TARVO, Inc.","subitem_text_language":"en"},{"subitem_text_value":"Information Technology Center, Nagoya University","subitem_text_language":"en"}]},"item_language":{"attribute_name":"言語","attribute_value_mlt":[{"subitem_language":"eng"}]},"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/228438/files/IPSJ-SLP23148008.pdf","label":"IPSJ-SLP23148008.pdf"},"format":"application/pdf","billing":["billing_file"],"filename":"IPSJ-SLP23148008.pdf","filesize":[{"value":"853.7 kB"}],"mimetype":"application/pdf","priceinfo":[{"tax":["include_tax"],"price":"0","billingrole":"22"},{"tax":["include_tax"],"price":"0","billingrole":"44"}],"accessrole":"open_login","version_id":"bd7feb05-04c8-4406-b0b4-60aa5b392be2","displaytype":"detail","licensetype":"license_note","license_note":"Copyright (c) 2023 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":"Lester, Phillip Violeta"}],"nameIdentifiers":[{}]},{"creatorNames":[{"creatorName":"Wen-ChinHuang, Ding Ma"}],"nameIdentifiers":[{}]},{"creatorNames":[{"creatorName":"山本, 龍一"}],"nameIdentifiers":[{}]},{"creatorNames":[{"creatorName":"小林, 和弘"}],"nameIdentifiers":[{}]},{"creatorNames":[{"creatorName":"戸田, 智基"}],"nameIdentifiers":[{}]}]},"item_4_creator_6":{"attribute_name":"著者名(英)","attribute_type":"creator","attribute_value_mlt":[{"creatorNames":[{"creatorName":"Lester, Phillip Violeta","creatorNameLang":"en"}],"nameIdentifiers":[{}]},{"creatorNames":[{"creatorName":"Wen-Chin, Huang","creatorNameLang":"en"}],"nameIdentifiers":[{}]},{"creatorNames":[{"creatorName":"Ding, Ma","creatorNameLang":"en"}],"nameIdentifiers":[{}]},{"creatorNames":[{"creatorName":"Ryuichi, Yamamoto","creatorNameLang":"en"}],"nameIdentifiers":[{}]},{"creatorNames":[{"creatorName":"Kazuhiro, Kobayashi","creatorNameLang":"en"}],"nameIdentifiers":[{}]},{"creatorNames":[{"creatorName":"Tomoki, Toda","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":"Although pretraining and fine-tuning approaches have proven to work well in speech intelligibility enhancement, various mismatches, such as the speech type mismatch or a speaker mismatches between the datasets used in each stage, can deteriorate the conversion performance of this framework. We propose a linguistic encoder robust enough to project both EL and typical speech in the same latent space, while still being able to extract accurate linguistic information, creating a unified representation to reduce the speech type mismatch. Furthermore, we introduce HuBERT output features to the proposed framework for reducing the speaker mismatch. Such a framework makes it possible to effectively use a large-scale parallel dataset during pretraining. We show that compared to the conventional framework using mel-spectrogram input and output features, using the proposed framework enables the model to synthesize more intelligible and naturally sounding speech, as shown by a significant 16% improvement in character error rate and 0.83 improvement in naturalness score.","subitem_description_type":"Other"}]},"item_4_description_8":{"attribute_name":"論文抄録(英)","attribute_value_mlt":[{"subitem_description":"Although pretraining and fine-tuning approaches have proven to work well in speech intelligibility enhancement, various mismatches, such as the speech type mismatch or a speaker mismatches between the datasets used in each stage, can deteriorate the conversion performance of this framework. We propose a linguistic encoder robust enough to project both EL and typical speech in the same latent space, while still being able to extract accurate linguistic information, creating a unified representation to reduce the speech type mismatch. Furthermore, we introduce HuBERT output features to the proposed framework for reducing the speaker mismatch. Such a framework makes it possible to effectively use a large-scale parallel dataset during pretraining. We show that compared to the conventional framework using mel-spectrogram input and output features, using the proposed framework enables the model to synthesize more intelligible and naturally sounding speech, as shown by a significant 16% improvement in character error rate and 0.83 improvement in naturalness score.","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":"2023-10-07","bibliographicIssueDateType":"Issued"},"bibliographicIssueNumber":"8","bibliographicVolumeNumber":"2023-SLP-148"}]},"relation_version_is_last":true,"weko_creator_id":"44499"},"created":"2025-01-19T01:27:35.148407+00:00","updated":"2025-01-19T11:51:08.016005+00:00"}