{"links":{},"id":57027,"metadata":{"_oai":{"id":"oai:ipsj.ixsq.nii.ac.jp:00057027","sets":["1164:5159:5180:5185"]},"path":["5185"],"owner":"1","recid":"57027","title":["講義の自動アーカイブ化のためのスライドと発話の対応付け"],"pubdate":{"attribute_name":"公開日","attribute_value":"2005-02-05"},"_buckets":{"deposit":"5b71c2b1-c4e6-4a17-8c6d-e6f0a889ae23"},"_deposit":{"id":"57027","pid":{"type":"depid","value":"57027","revision_id":0},"owners":[1],"status":"published","created_by":1},"item_title":"講義の自動アーカイブ化のためのスライドと発話の対応付け","author_link":["0","0"],"item_titles":{"attribute_name":"タイトル","attribute_value_mlt":[{"subitem_title":"講義の自動アーカイブ化のためのスライドと発話の対応付け"},{"subitem_title":"Automatic Alignment of Speech Transcriptions with Viewgraph Slides for Lecture Archiving","subitem_title_language":"en"}]},"item_type_id":"4","publish_date":"2005-02-05","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 知能情報学専攻"},{"subitem_text_value":"京都大学 情報学研究科 知能情報学専攻"}]},"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"}]},"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/57027/files/IPSJ-SLP04055011.pdf"},"date":[{"dateType":"Available","dateValue":"2007-02-05"}],"format":"application/pdf","billing":["billing_file"],"filename":"IPSJ-SLP04055011.pdf","filesize":[{"value":"414.7 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":"22"},{"tax":["include_tax"],"price":"0","billingrole":"44"}],"accessrole":"open_date","version_id":"9b3bdba5-c57a-454c-bccb-b13496257519","displaytype":"detail","licensetype":"license_note","license_note":"Copyright (c) 2005 by the Information Processing Society of Japan"}]},"item_4_creator_5":{"attribute_name":"著者名","attribute_type":"creator","attribute_value_mlt":[{"creatorNames":[{"creatorName":"北出, 祐"},{"creatorName":"河原, 達也"}],"nameIdentifiers":[{}]}]},"item_4_creator_6":{"attribute_name":"著者名(英)","attribute_type":"creator","attribute_value_mlt":[{"creatorNames":[{"creatorName":"Tasuku, Kitade","creatorNameLang":"en"},{"creatorName":"Tatsuya, Kawahara","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":"講義の自動アーカイブ化を目的として，講義音声の書き起こしの各発話と講義に用いられたスライドを自動的に対応付ける手法を提案する．スライドのテキストからキーワードを抽出し，発話との関連度を定義した．その際に単一のスライドへの対応付けが困難である場合もあるので，複数のスライドからなるトピック単位も構成した．また，発話直前のポーズ長や談話標識の出現に着目し，話題の遷移尤度を定義した．これらの尺度を用いてスライドまたはトピックを状態とするマルコフモデルを構築し，スライドやトピック単位に発話の対応付けを行った．実際の大学の講義に対して自動対応付けの実験を行ったところ，F値が0.681であった．","subitem_description_type":"Other"}]},"item_4_description_8":{"attribute_name":"論文抄録(英)","attribute_value_mlt":[{"subitem_description":"Automatic alignment of speech transcriptions with a sequence of viewgraph slides used in a lecture is addressed. We extracted the keywords from the text of the slides and define the similarity with utterances. Here, we introduce a topic unit consisting of multimple slides for more stable and flexible matching. In addition, we use the information of pauses and discourse markers for defining the likelihood of transition between slides or topics. Based on these measures, we set up a Markov model to be matched with utterances. Experimental results using two lectures confirm the effectiveness of the method.","subitem_description_type":"Other"}]},"item_4_biblio_info_10":{"attribute_name":"書誌情報","attribute_value_mlt":[{"bibliographicPageEnd":"64","bibliographic_titles":[{"bibliographic_title":"情報処理学会研究報告音声言語情報処理（SLP）"}],"bibliographicPageStart":"59","bibliographicIssueDates":{"bibliographicIssueDate":"2005-02-05","bibliographicIssueDateType":"Issued"},"bibliographicIssueNumber":"12(2004-SLP-055)","bibliographicVolumeNumber":"2005"}]},"relation_version_is_last":true,"weko_creator_id":"1"},"created":"2025-01-18T23:20:18.994902+00:00","updated":"2025-01-22T04:41:44.602368+00:00"}