{"links":{},"id":79345,"metadata":{"_oai":{"id":"oai:ipsj.ixsq.nii.ac.jp:00079345","sets":["1164:5159:6316:6625"]},"path":["6625"],"owner":"10","recid":"79345","title":["識別学習に基づく音声認識単語辞書の最適化の評価"],"pubdate":{"attribute_name":"公開日","attribute_value":"2011-12-12"},"_buckets":{"deposit":"e7c57273-1455-4a3a-9aa6-a319a1bce008"},"_deposit":{"id":"79345","pid":{"type":"depid","value":"79345","revision_id":0},"owners":[10],"status":"published","created_by":10},"item_title":"識別学習に基づく音声認識単語辞書の最適化の評価","author_link":["0","0"],"item_titles":{"attribute_name":"タイトル","attribute_value_mlt":[{"subitem_title":"識別学習に基づく音声認識単語辞書の最適化の評価"},{"subitem_title":"Evaluation of Lexicon Optimization based on Discriminative Learning","subitem_title_language":"en"}]},"item_keyword":{"attribute_name":"キーワード","attribute_value_mlt":[{"subitem_subject":"言語モデル・辞書","subitem_subject_scheme":"Other"}]},"item_type_id":"4","publish_date":"2011-12-12","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／新疆大学信息学院"},{"subitem_text_value":"京都大学情報学研究科"},{"subitem_text_value":"新疆大学信息学院"}]},"item_4_text_4":{"attribute_name":"著者所属(英)","attribute_value_mlt":[{"subitem_text_value":"School of Informatics, Kyoto University, Kyoto, Japan / Institute of Information Engineering, Xinjiang University, Urumqi, China","subitem_text_language":"en"},{"subitem_text_value":"School of Informatics, Kyoto University, Kyoto, Japan","subitem_text_language":"en"},{"subitem_text_value":"Institute of Information Engineering, Xinjiang University, Urumqi, China","subitem_text_language":"en"}]},"item_language":{"attribute_name":"言語","attribute_value_mlt":[{"subitem_language":"eng"}]},"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/79345/files/IPSJ-SLP11089002.pdf"},"date":[{"dateType":"Available","dateValue":"2013-12-12"}],"format":"application/pdf","billing":["billing_file"],"filename":"IPSJ-SLP11089002.pdf","filesize":[{"value":"268.5 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":"cffb3489-cf14-4f4a-ba61-43854ce39a9d","displaytype":"detail","licensetype":"license_note","license_note":"Copyright (c) 2011 by the Information Processing Society of Japan"}]},"item_4_creator_5":{"attribute_name":"著者名","attribute_type":"creator","attribute_value_mlt":[{"creatorNames":[{"creatorName":"アブリミテ・ミジテ"},{"creatorName":"河原, 達也"},{"creatorName":"ハムヅラ・アスカ"}],"nameIdentifiers":[{}]}]},"item_4_creator_6":{"attribute_name":"著者名(英)","attribute_type":"creator","attribute_value_mlt":[{"creatorNames":[{"creatorName":"Mijit, Ablimit","creatorNameLang":"en"},{"creatorName":"Tatsuya, Kawahara","creatorNameLang":"en"},{"creatorName":"Askar, Hamdulla","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":"日本語やウイグル語のような膠着言語では、単語の単位の定義が自明でない。音声認識の単位として形態素が一般的に用いられるが、多くの形態素が短いため、制約として弱く、誤認識の原因になる。我々は、識別学習に基づいて、音声認識誤りの削減に直結するような単語のエントリを選択する方法を提案している。本手法では、各単語エントリに対して、素性の集合とそれらの重みからなる評価関数、及び、形態素単位のモデルと単語単位のモデルの誤り率の差による誤分類尺度を定義した上で、パーセプトロン学習によって素性の重みを学習する。その上で、評価関数の値の高い単語もしくはサブワードのエントリを辞書に追加する。本手法をウイグル語の大語彙連続音声認識システムに適用し、(1) 有効な素性、(2) サブワードエントリの効果、に関して評価を行った。さらに、従来の頻度や相互情報量に基づく手法と比較した上で、それらとの組み合わせも検討した。","subitem_description_type":"Other"}]},"item_4_description_8":{"attribute_name":"論文抄録(英)","attribute_value_mlt":[{"subitem_description":"In agglutinative languages, selection of lexical unit is not obvious. Morpheme unit is usually adopted to ensure a sufficient coverage, but many morphemes are short, resulting in weak constraints and possible confusions. In this paper, we propose a discriminative approach to select lexical entries which will directly contribute to ASR error reduction. We define an evaluation function for each word by a set of features and their weights, and the measure for optimization by the difference of WERs by the morpheme-based model and by the word-based model. Then, the weights of the features are learned by a perceptron algorithm. Finally, word (or sub-word) entries with higher evaluation scores are selected to be added to the lexicon. This method is successfully applied to an Uyghur large-vocabulary continuous speech recognition system, resulting in a significant reduction of WER and the lexicon size. Further improvement is achieved by combining with a statistical method based on mutual information criterion.","subitem_description_type":"Other"}]},"item_4_biblio_info_10":{"attribute_name":"書誌情報","attribute_value_mlt":[{"bibliographicPageEnd":"5","bibliographic_titles":[{"bibliographic_title":"研究報告音声言語情報処理（SLP）"}],"bibliographicPageStart":"1","bibliographicIssueDates":{"bibliographicIssueDate":"2011-12-12","bibliographicIssueDateType":"Issued"},"bibliographicIssueNumber":"2","bibliographicVolumeNumber":"2011-SLP-89"}]},"relation_version_is_last":true,"weko_creator_id":"10"},"created":"2025-01-18T23:34:06.441065+00:00","updated":"2025-01-21T20:15:20.807498+00:00"}