{"links":{},"metadata":{"_oai":{"id":"oai:ipsj.ixsq.nii.ac.jp:00075428","sets":["1164:5159:6316:6483"]},"path":["6483"],"owner":"10","recid":"75428","title":["識別学習に基づく音声認識単語辞書の最適化"],"pubdate":{"attribute_name":"公開日","attribute_value":"2011-07-14"},"_buckets":{"deposit":"5bebf7b2-b071-4fda-961a-5921a41d6554"},"_deposit":{"id":"75428","pid":{"type":"depid","value":"75428","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","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"},{"subitem_title":"Lexicon Optimization for Automatic Speech Recognition based on Discriminative Learning","subitem_title_language":"en"}]},"item_keyword":{"attribute_name":"キーワード","attribute_value_mlt":[{"subitem_subject":"音声ドキュメント・検索","subitem_subject_scheme":"Other"}]},"item_type_id":"4","publish_date":"2011-07-14","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"},{"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":"Kyoto University, School of Informatics","subitem_text_language":"en"},{"subitem_text_value":"Kyoto University, School of Informatics","subitem_text_language":"en"},{"subitem_text_value":"Xinjiang University, Information Institute","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/75428/files/IPSJ-SLP11087005.pdf"},"date":[{"dateType":"Available","dateValue":"2013-07-14"}],"format":"application/pdf","billing":["billing_file"],"filename":"IPSJ-SLP11087005.pdf","filesize":[{"value":"232.7 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":"abf8c327-2e02-40c9-ba18-7f04e9535c25","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、単語の単位が自明でなく、音声認識の言語モデルの設計においても重要な問題となっている。本稿では、音声認識誤り （単語誤り率） を削減するような単語エントリを識別学習により選択する方法を提案する。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。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、形態素単位のモデルに比べて語彙サイズをあまり増やすことなく、単語誤り率を大きく削減することができた。","subitem_description_type":"Other"}]},"item_4_description_8":{"attribute_name":"論文抄録(英)","attribute_value_mlt":[{"subitem_description":"In agglutinative languages such as Japanese and Uyghur, selection of lexical unit is not obvious and one of the important issues in designing language model for automatic speech recognition (ASR). In this paper, we propose a discriminative learning method to select word entries which would reduce the word error rate (WER). 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 two units (morpheme and word). Then, the weights of the features are learned by a perceptron algorithm. Finally, word entries with higher evaluation scores are selected. The discriminative method is successfully applied to an Uyghur large-vocabulary continuous speech recognition system, resulting in a significant reduction of WER without a drastic increase of the vocabulary size.","subitem_description_type":"Other"}]},"item_4_biblio_info_10":{"attribute_name":"書誌情報","attribute_value_mlt":[{"bibliographicPageEnd":"4","bibliographic_titles":[{"bibliographic_title":"研究報告音声言語情報処理（SLP）"}],"bibliographicPageStart":"1","bibliographicIssueDates":{"bibliographicIssueDate":"2011-07-14","bibliographicIssueDateType":"Issued"},"bibliographicIssueNumber":"5","bibliographicVolumeNumber":"2011-SLP-87"}]},"relation_version_is_last":true,"weko_creator_id":"10"},"updated":"2025-01-21T21:14:26.673241+00:00","created":"2025-01-18T23:32:31.773612+00:00","id":75428}