{"created":"2025-01-18T23:20:00.876910+00:00","links":{},"updated":"2025-01-22T04:52:38.494149+00:00","id":56640,"metadata":{"_oai":{"id":"oai:ipsj.ixsq.nii.ac.jp:00056640","sets":["1164:5159:5162:5165"]},"path":["5165"],"owner":"1","recid":"56640","title":["音声認識のための言語処理 ： 何が足りないか？"],"pubdate":{"attribute_name":"公開日","attribute_value":"2008-07-11"},"_buckets":{"deposit":"2028f99b-b05e-4d9c-bdc0-32fce4e12efd"},"_deposit":{"id":"56640","pid":{"type":"depid","value":"56640","revision_id":0},"owners":[1],"status":"published","created_by":1},"item_title":"音声認識のための言語処理 ： 何が足りないか？","author_link":["0","0"],"item_titles":{"attribute_name":"タイトル","attribute_value_mlt":[{"subitem_title":"音声認識のための言語処理 ： 何が足りないか？"},{"subitem_title":"Language Processing for Speech Recognition : What is Missing?","subitem_title_language":"en"}]},"item_type_id":"4","publish_date":"2008-07-11","item_4_text_3":{"attribute_name":"著者所属","attribute_value_mlt":[{"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":"Academic Center for Computing and Media Studies","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/56640/files/IPSJ-SLP08072007.pdf"},"date":[{"dateType":"Available","dateValue":"2010-07-11"}],"format":"application/pdf","billing":["billing_file"],"filename":"IPSJ-SLP08072007.pdf","filesize":[{"value":"1.6 MB"}],"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":"c4f14370-5073-4d9c-9195-0f26ad9a2d14","displaytype":"detail","licensetype":"license_note","license_note":"Copyright (c) 2008 by the Information Processing Society of Japan"}]},"item_4_creator_5":{"attribute_name":"著者名","attribute_type":"creator","attribute_value_mlt":[{"creatorNames":[{"creatorName":"森, 信介"}],"nameIdentifiers":[{}]}]},"item_4_creator_6":{"attribute_name":"著者名(英)","attribute_type":"creator","attribute_value_mlt":[{"creatorNames":[{"creatorName":"Shinsuke, Mori","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":"言語モデルとしてよく用いられる単語 n-gram モデルは、適切に単語に分割された対象分野のコーパスが大量に利用可能な状況では実用上十分な性能を実現できる。しかし、対象分野の言語資源としては、単語境界情報のない生コーパスだけが利用可能であったり、基準に一貫性のない単語 (表記) の一覧が与えらえるのみであるという状況が現実的にはほとんどである。本稿では、一般的な分野の単語分割済みコーパスとそこに出現する単語の発音辞書があり、ある程度の量の対象分野の生コーパスが利用可能であるとの前提の下、音声認識のための言語モデルを効率よく構築することを目的として、現在利用可能な言語処理技術について述べるとともに、望まれる言語処理技術について考察する。","subitem_description_type":"Other"}]},"item_4_description_8":{"attribute_name":"論文抄録(英)","attribute_value_mlt":[{"subitem_description":"A word n-gram model, which is often used as a language model, functions sufficiently well for practical uses under the condition that a large corpus in the target domain with word boundary information is available. In many practial cases, however, only a raw corpus without word boundary information is available, or a list of spellings of words and word sequences is given. In this article, assuming that we have a general corpus with word boundary information, a word list with pronunciation appearing in that corpus, and a certain amount of a raw corpus in the target domain, we discuss current natural language processing technologies and missing ones.","subitem_description_type":"Other"}]},"item_4_biblio_info_10":{"attribute_name":"書誌情報","attribute_value_mlt":[{"bibliographicPageEnd":"42","bibliographic_titles":[{"bibliographic_title":"情報処理学会研究報告音声言語情報処理（SLP）"}],"bibliographicPageStart":"37","bibliographicIssueDates":{"bibliographicIssueDate":"2008-07-11","bibliographicIssueDateType":"Issued"},"bibliographicIssueNumber":"68(2008-SLP-072)","bibliographicVolumeNumber":"2008"}]},"relation_version_is_last":true,"weko_creator_id":"1"}}