{"links":{},"id":57198,"metadata":{"_oai":{"id":"oai:ipsj.ixsq.nii.ac.jp:00057198","sets":["1164:5159:5192:5193"]},"path":["5193"],"owner":"1","recid":"57198","title":["2パス探索アルゴリズムにおける高速な単語事後確率に基づく信頼度算出法"],"pubdate":{"attribute_name":"公開日","attribute_value":"2003-12-19"},"_buckets":{"deposit":"a168ce21-f100-43b3-9aa2-cbc02413f01c"},"_deposit":{"id":"57198","pid":{"type":"depid","value":"57198","revision_id":0},"owners":[1],"status":"published","created_by":1},"item_title":"2パス探索アルゴリズムにおける高速な単語事後確率に基づく信頼度算出法","author_link":["0","0"],"item_titles":{"attribute_name":"タイトル","attribute_value_mlt":[{"subitem_title":"2パス探索アルゴリズムにおける高速な単語事後確率に基づく信頼度算出法"},{"subitem_title":"Real - Time Confidence Scoring Based on Word Posterior Probability on two - pass search algorithm","subitem_title_language":"en"}]},"item_type_id":"4","publish_date":"2003-12-19","item_4_text_3":{"attribute_name":"著者所属","attribute_value_mlt":[{"subitem_text_value":"奈良先端科学技術大学院大学情報科学研究科"},{"subitem_text_value":"京都大学学術情報メディアセンター"},{"subitem_text_value":"奈良先端科学技術大学院大学情報科学研究科"}]},"item_4_text_4":{"attribute_name":"著者所属(英)","attribute_value_mlt":[{"subitem_text_value":"Graduate School of Information Science, Nara Institute of Science and Technology","subitem_text_language":"en"},{"subitem_text_value":"School of Informatics, Kyoto University","subitem_text_language":"en"},{"subitem_text_value":"Graduate School of Information Science, Nara Institute of Science and 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/57198/files/IPSJ-SLP03049048.pdf"},"date":[{"dateType":"Available","dateValue":"2005-12-19"}],"format":"application/pdf","billing":["billing_file"],"filename":"IPSJ-SLP03049048.pdf","filesize":[{"value":"638.2 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":"250e0fbd-c42d-4c7d-8e84-303e19391031","displaytype":"detail","licensetype":"license_note","license_note":"Copyright (c) 2003 by the Information Processing Society of Japan"}]},"item_4_creator_5":{"attribute_name":"著者名","attribute_type":"creator","attribute_value_mlt":[{"creatorNames":[{"creatorName":"李晃伸"},{"creatorName":"川原, 達也"},{"creatorName":"鹿野, 清宏"}],"nameIdentifiers":[{}]}]},"item_4_creator_6":{"attribute_name":"著者名(英)","attribute_type":"creator","attribute_value_mlt":[{"creatorNames":[{"creatorName":"Akinobu, Lee","creatorNameLang":"en"},{"creatorName":"Tatsuya, Kawahara","creatorNameLang":"en"},{"creatorName":"Kiyohiro, Shikano","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":"音声認識システムにおいて，認識結果に対して事後確率などを用いて信頼度を付与することで，発話検証や対話管理などの音声アプリケーションにおいて認識誤りを考慮したより高度な処理を行うことができる．この単語の事後確率を用いた信頼度算出では，通常，認識処理（デコーディング）の結果得られた仮説群のゆう度をもとに計算されるが，十分な精度の確信度を得るためにはN?best候補で数百以上の大量の文仮説を求める必要があり，多くの計算量を必要とする．本研究では，２パストリートレリス検索に基づくデコーディングにおいて，検索中に得られる部分文化説の尤度から単語の信頼度を簡易かつ高速に算出するアルゴリズムを提案する．後段パスのスタックデコーディングにおける単語仮説展開時に，その次単語仮説の集合およびそれぞれから展開される新たな仮説のヒューリスティックを含む尤度から，その展開単語の事後確率を計算する．通常のデコーディング処理に対して極めて少ない計算量で信頼度を計算できる．認識エンジンJuliusにおいて，N-best候補から事後確率を算出する従来手法との比較を行った結果，提案手法は大量のN-best候補を求める必要がないことから認識処理全体を非常に高速に行え，また信頼度の精度も，簡易な計算法ながらN?best候補を用いる手法と同等以上の信頼度を算出することが示された．","subitem_description_type":"Other"}]},"item_4_description_8":{"attribute_name":"論文抄録(英)","attribute_value_mlt":[{"subitem_description":"Confidence scoring based on word posterior probability is usually performed as a post process of speech recognition decoding, and also needs a large number of word hypotheses to get enough confidence quality. We propose a imple way of computing the word confidence using estimated posterior probability while decoding. At the word expansion of stack decoding search, the local sentence likehoods that contains heuristic scores of unreached segment are directly used to compute the posterior probabilities. Experimental result showed that, although the likelihoods are not optimal, it can provide slightly better confidence measures compared with N-best lists, while the computation is much faster because no N-best decoding is required.","subitem_description_type":"Other"}]},"item_4_biblio_info_10":{"attribute_name":"書誌情報","attribute_value_mlt":[{"bibliographicPageEnd":"286","bibliographic_titles":[{"bibliographic_title":"情報処理学会研究報告音声言語情報処理（SLP）"}],"bibliographicPageStart":"281","bibliographicIssueDates":{"bibliographicIssueDate":"2003-12-19","bibliographicIssueDateType":"Issued"},"bibliographicIssueNumber":"124(2003-SLP-049)","bibliographicVolumeNumber":"2003"}]},"relation_version_is_last":true,"weko_creator_id":"1"},"created":"2025-01-18T23:20:27.194750+00:00","updated":"2025-01-22T04:37:46.784049+00:00"}