{"links":{},"id":57532,"metadata":{"_oai":{"id":"oai:ipsj.ixsq.nii.ac.jp:00057532","sets":["1164:5159:5210:5214"]},"path":["5214"],"owner":"1","recid":"57532","title":["種々のタスクにおける大語彙連続音声認識システムの性能評価と診断"],"pubdate":{"attribute_name":"公開日","attribute_value":"2000-06-02"},"_buckets":{"deposit":"097fe115-55bb-4b98-adba-3a40b2ac5994"},"_deposit":{"id":"57532","pid":{"type":"depid","value":"57532","revision_id":0},"owners":[1],"status":"published","created_by":1},"item_title":"種々のタスクにおける大語彙連続音声認識システムの性能評価と診断","author_link":["0","0"],"item_titles":{"attribute_name":"タイトル","attribute_value_mlt":[{"subitem_title":"種々のタスクにおける大語彙連続音声認識システムの性能評価と診断"},{"subitem_title":"Diagnosis and evaluation of various LVCSR systems","subitem_title_language":"en"}]},"item_type_id":"4","publish_date":"2000-06-02","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"},{"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"},{"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":"Graduate School of Informatics Kyoto University","subitem_text_language":"en"},{"subitem_text_value":"Graduate School of Informatics Kyoto University","subitem_text_language":"en"},{"subitem_text_value":"Graduate School of Informatics Kyoto University","subitem_text_language":"en"},{"subitem_text_value":"Graduate School of Informatics Kyoto University","subitem_text_language":"en"},{"subitem_text_value":"Graduate 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/57532/files/IPSJ-SLP00031011.pdf"},"date":[{"dateType":"Available","dateValue":"2002-06-02"}],"format":"application/pdf","billing":["billing_file"],"filename":"IPSJ-SLP00031011.pdf","filesize":[{"value":"875.7 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":"9df4c999-5a28-4cd5-9c8d-76993e39d1da","displaytype":"detail","licensetype":"license_note","license_note":"Copyright (c) 2000 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":"加藤, 一臣"},{"creatorName":"三村, 正人"},{"creatorName":"李晃伸"},{"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":"Kazuomi, Kato","creatorNameLang":"en"},{"creatorName":"Masato, Mimura","creatorNameLang":"en"},{"creatorName":"Akinobu, Lee","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":"大規模な統計的モデルに基づく大語彙連続音声認識における認識誤りの原因を自動的に同定・診断する手法を，読上げ音声，講演音声，対話音声の三つの異なる認識システムに適用し，現状の音声認識システムの評価を行う．この手法は，正解文と認識結果のスコアを比較し，区間と要因ごとに分解するものであり，誤り単語を含む区間ごとに対して，原因モジュールを音響モデル・言語モデル・デコーダのいずれかに同定し，その傾向を分析する．読上げ音声認識システムにおいては，サーチエラーを自動同定・分析することにより，実際に効率的にデコーダの改善を行うことができた．講演や対話のような話し言葉の音声認識では，間投詞やショートポーズに関する言語モデルや，典型的なつなぎ語や文末表現における音素コンテクストのモデルに問題があることが明らかになった．また，同一の発声スタイルのデータで音響モデル・言語モデルを構築することの重要性・有効性が確認された．","subitem_description_type":"Other"}]},"item_4_description_8":{"attribute_name":"論文抄録(英)","attribute_value_mlt":[{"subitem_description":"We have proposed a method to automatically diagnose recognition errors in large vocabulary continuous speech recognition. For every erroneous segment, it identifies the error-causing module among acoustic model, language model and the decoder. More detailed diagnosis is also output. This method has been applied to three LVCSR systems: read speech, lecture speech and dialogue speech recognition system. In the read speech recognition system, the diagnosis is useful to improve our decoder Julius. In the lecture and dialogue recognition systems, we have made clear the significance of matching training database in its speaking style to the task.","subitem_description_type":"Other"}]},"item_4_biblio_info_10":{"attribute_name":"書誌情報","attribute_value_mlt":[{"bibliographicPageEnd":"80","bibliographic_titles":[{"bibliographic_title":"情報処理学会研究報告音声言語情報処理（SLP）"}],"bibliographicPageStart":"73","bibliographicIssueDates":{"bibliographicIssueDate":"2000-06-02","bibliographicIssueDateType":"Issued"},"bibliographicIssueNumber":"54(2000-SLP-031)","bibliographicVolumeNumber":"2000"}]},"relation_version_is_last":true,"weko_creator_id":"1"},"created":"2025-01-18T23:20:42.921716+00:00","updated":"2025-01-22T04:27:49.081328+00:00"}