{"created":"2025-01-18T23:20:21.418926+00:00","links":{},"updated":"2025-01-22T04:41:00.332755+00:00","id":57078,"metadata":{"_oai":{"id":"oai:ipsj.ixsq.nii.ac.jp:00057078","sets":["1164:5159:5186:5187"]},"path":["5187"],"owner":"1","recid":"57078","title":["音声理解のための音声認識評価尺度とベイズリスク最小化デコーディング"],"pubdate":{"attribute_name":"公開日","attribute_value":"2004-12-22"},"_buckets":{"deposit":"d3e2182d-a8b2-4f92-9366-6df21a77c5f7"},"_deposit":{"id":"57078","pid":{"type":"depid","value":"57078","revision_id":0},"owners":[1],"status":"published","created_by":1},"item_title":"音声理解のための音声認識評価尺度とベイズリスク最小化デコーディング","author_link":["0","0"],"item_titles":{"attribute_name":"タイトル","attribute_value_mlt":[{"subitem_title":"音声理解のための音声認識評価尺度とベイズリスク最小化デコーディング"},{"subitem_title":"ASR Evaluation Measure and Minimum Bayes - Risk Decoding for Open - domain Speech Understanding","subitem_title_language":"en"}]},"item_type_id":"4","publish_date":"2004-12-22","item_4_text_3":{"attribute_name":"著者所属","attribute_value_mlt":[{"subitem_text_value":"龍谷大学理工学部情報メディア学科"},{"subitem_text_value":"京都大学学術情報メディアセンター"}]},"item_4_text_4":{"attribute_name":"著者所属(英)","attribute_value_mlt":[{"subitem_text_value":"Faculty of Science and Technology, Ryukoku University","subitem_text_language":"en"},{"subitem_text_value":"Academic Center for Computing and Media Studies, 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/57078/files/IPSJ-SLP04054042.pdf"},"date":[{"dateType":"Available","dateValue":"2006-12-22"}],"format":"application/pdf","billing":["billing_file"],"filename":"IPSJ-SLP04054042.pdf","filesize":[{"value":"829.6 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":"5db79c3e-41a9-4cf3-a74e-ac1bf22e1709","displaytype":"detail","licensetype":"license_note","license_note":"Copyright (c) 2004 by the Information Processing Society of Japan"}]},"item_4_creator_5":{"attribute_name":"著者名","attribute_type":"creator","attribute_value_mlt":[{"creatorNames":[{"creatorName":"南條浩輝"},{"creatorName":"河原, 達也"}],"nameIdentifiers":[{}]}]},"item_4_creator_6":{"attribute_name":"著者名(英)","attribute_type":"creator","attribute_value_mlt":[{"creatorNames":[{"creatorName":"Hiroaki, Nanjo","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":"ドメインを限定しない自然な話し言葉の音声理解を目的とした音声認識の評価尺度とそれに基づくデコーディング手法を提案する．従来，音声認識の一般的な評価尺度として，全ての単語を一様に扱う「単語誤り率(word error rate: WER)」が用いられてきた．これに対して，情報検索の観点から各単語の重要度を考慮した「重みつきキーワード」誤り率(weighted keyword error rate: WKER)」を提案する．講演音声からの重要文抽出のタスクにおいて，重みつきキーワード誤り率が重要文抽出の制度と相関が高いことを示す．その上で，ベイズリスク最小化(Minimum Bayes-Risk: MBR)」の枠組みに基づいて，重みつきキーワード誤り率の最小化を行う音声認識を実現する．CSJの学会講演17講演を用いて評価を行い，提案する認識手法が重みつきキーワード誤り率及び重要文抽出精度の改善に効果があることを示す．","subitem_description_type":"Other"}]},"item_4_description_8":{"attribute_name":"論文抄録(英)","attribute_value_mlt":[{"subitem_description":"A new evaluation measure of speech recognition and a decoding strategy for keyword-based open-domain speech understanding are presented. Conventionally, WER (word error rate) has been widely used as an evaluation measure of speech recognition, which treats all words in a uniform manner. In this paper, we define a weighted keyword error rate (WKER) which gives a weight on errors from a viewpoint of information retrieval. We first demonstrate that this measure is more appropriate for predicting the performance of key sentence indexing of oral presentations. Then, we formulate a decoding method to minimize WKER based on Minimum Bayes-Risk (MBR) framework, and show that the decoding method works reasonably for improving WKER and key sentence indexing.","subitem_description_type":"Other"}]},"item_4_biblio_info_10":{"attribute_name":"書誌情報","attribute_value_mlt":[{"bibliographicPageEnd":"252","bibliographic_titles":[{"bibliographic_title":"情報処理学会研究報告音声言語情報処理（SLP）"}],"bibliographicPageStart":"247","bibliographicIssueDates":{"bibliographicIssueDate":"2004-12-22","bibliographicIssueDateType":"Issued"},"bibliographicIssueNumber":"131(2004-SLP-054)","bibliographicVolumeNumber":"2004"}]},"relation_version_is_last":true,"weko_creator_id":"1"}}