{"links":{},"id":56666,"metadata":{"_oai":{"id":"oai:ipsj.ixsq.nii.ac.jp:00056666","sets":["1164:5159:5162:5166"]},"path":["5166"],"owner":"1","recid":"56666","title":["話し言葉の整形作業における削除箇所の自動同定"],"pubdate":{"attribute_name":"公開日","attribute_value":"2008-05-15"},"_buckets":{"deposit":"1886a5b1-4aa8-4b7f-bc50-a5fd44aeed66"},"_deposit":{"id":"56666","pid":{"type":"depid","value":"56666","revision_id":0},"owners":[1],"status":"published","created_by":1},"item_title":"話し言葉の整形作業における削除箇所の自動同定","author_link":["0","0"],"item_titles":{"attribute_name":"タイトル","attribute_value_mlt":[{"subitem_title":"話し言葉の整形作業における削除箇所の自動同定"},{"subitem_title":"Automatic Detection of Portions to be Deleted during Cleaning Transcripts","subitem_title_language":"en"}]},"item_type_id":"4","publish_date":"2008-05-15","item_4_text_3":{"attribute_name":"著者所属","attribute_value_mlt":[{"subitem_text_value":"京都大学情報学研究科"},{"subitem_text_value":"京都大学情報学研究科"},{"subitem_text_value":"京都大学情報学研究科"},{"subitem_text_value":"情報通信研究機構"}]},"item_4_text_4":{"attribute_name":"著者所属(英)","attribute_value_mlt":[{"subitem_text_value":"School of Informatics, Kyoto University","subitem_text_language":"en"},{"subitem_text_value":"School of Informatics, Kyoto University","subitem_text_language":"en"},{"subitem_text_value":"School of Informatics, Kyoto University","subitem_text_language":"en"},{"subitem_text_value":"National Institute of Information and Communications Technology","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/56666/files/IPSJ-SLP08071013.pdf"},"date":[{"dateType":"Available","dateValue":"2010-05-15"}],"format":"application/pdf","billing":["billing_file"],"filename":"IPSJ-SLP08071013.pdf","filesize":[{"value":"1.2 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":"1b4ea33a-d84e-4d1a-b745-b9761189b89e","displaytype":"detail","licensetype":"license_note","license_note":"Copyright (c) 2008 by the Information Processing Society of Japan"}]},"item_4_creator_5":{"attribute_name":"著者名","attribute_type":"creator","attribute_value_mlt":[{"creatorNames":[{"creatorName":"尾嶋, 憲治"},{"creatorName":"河原, 達也"},{"creatorName":"秋田, 祐哉"},{"creatorName":"内元, 清貴"}],"nameIdentifiers":[{}]}]},"item_4_creator_6":{"attribute_name":"著者名(英)","attribute_type":"creator","attribute_value_mlt":[{"creatorNames":[{"creatorName":"Kenji, Ojima","creatorNameLang":"en"},{"creatorName":"Tatsuya, Kawahara","creatorNameLang":"en"},{"creatorName":"Yuya, Akita","creatorNameLang":"en"},{"creatorName":"Kiyotaka, Uchimoto","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_description_7":{"attribute_name":"論文抄録","attribute_value_mlt":[{"subitem_description":"『日本語話し言葉コーパス』 （CSJ） のコアに対して、講演録の作成を想定して、削除・置換・挿入などの一次整形のアノテーションを付与した。本稿では、この話し言葉の整形作業において削除される箇所を自動推定する手法を提案する。本研究では、削除箇所について IOB 体系でラベリングを行い、このラベルについてのテキストチャンキングの問題とみなし、SVM に基づくチャンカである Yam Cha を利用する。素性として、同一表現の繰り返しや文節・係り受けといった言語的情報、基本周波数などの韻律的情報などを検討し、また言い直しなどの自己修復部との関係についても調べた。書き起こしと音声認識結果の両方に対して評価実験を行い、書き起こしについては F 値 0.8 が得られた。","subitem_description_type":"Other"}]},"item_4_description_8":{"attribute_name":"論文抄録(英)","attribute_value_mlt":[{"subitem_description":"We introduce a corpus annotated with editing operations (deletion, insertion, substitution) on faithful speech transcripts for generating document-style texts, and propose a method to automatically detect portions to be deleted during this process. The proposed method is based on an SVM chunker. Features for the SVM include linguistic information such as existence of repeated utterances and syntactic relations as well as prosodic information. We achieved reasonable results not only for the manual transcripts but also for ASR results. ","subitem_description_type":"Other"}]},"item_4_biblio_info_10":{"attribute_name":"書誌情報","attribute_value_mlt":[{"bibliographicPageEnd":"91","bibliographic_titles":[{"bibliographic_title":"情報処理学会研究報告音声言語情報処理（SLP）"}],"bibliographicPageStart":"85","bibliographicIssueDates":{"bibliographicIssueDate":"2008-05-15","bibliographicIssueDateType":"Issued"},"bibliographicIssueNumber":"46(2008-SLP-071)","bibliographicVolumeNumber":"2008"}]},"relation_version_is_last":true,"weko_creator_id":"1"},"created":"2025-01-18T23:20:02.085322+00:00","updated":"2025-01-22T04:52:29.157986+00:00"}