{"created":"2025-01-18T23:20:32.091495+00:00","metadata":{"_oai":{"id":"oai:ipsj.ixsq.nii.ac.jp:00057303","sets":["1164:5159:5198:5199"]},"path":["5199"],"owner":"1","recid":"57303","title":["討論音声を対象とした話者モデル選択による話者インデキシングと自動書き起こし"],"pubdate":{"attribute_name":"公開日","attribute_value":"2002-12-16"},"_buckets":{"deposit":"e1740169-83c4-4753-82bd-ea87f5a868a5"},"_deposit":{"id":"57303","pid":{"type":"depid","value":"57303","revision_id":0},"owners":[1],"status":"published","created_by":1},"item_title":"討論音声を対象とした話者モデル選択による話者インデキシングと自動書き起こし","author_link":["0","0"],"item_titles":{"attribute_name":"タイトル","attribute_value_mlt":[{"subitem_title":"討論音声を対象とした話者モデル選択による話者インデキシングと自動書き起こし"},{"subitem_title":"Speaker Indexing based on Speaker Model Selection and Automatic Speech Recognition in Discussions","subitem_title_language":"en"}]},"item_type_id":"4","publish_date":"2002-12-16","item_4_text_3":{"attribute_name":"著者所属","attribute_value_mlt":[{"subitem_text_value":"科学技術振興事業団さきがけ研究21「協調と制御」領域"},{"subitem_text_value":"科学技術振興事業団さきがけ研究21「協調と制御」領域／京都大学情報学研究科知能情報学専攻"},{"subitem_text_value":"科学技術振興事業団さきがけ研究21「協調と制御」領域／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":"PRESTO, Japan Science and Technology Corporation (JST)","subitem_text_language":"en"},{"subitem_text_value":"PRESTO, Japan Science and Technology Corporation (JST)/School of Informatics, Kyoto University","subitem_text_language":"en"},{"subitem_text_value":"PRESTO, Japan Science and Technology Corporation (JST)/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/57303/files/IPSJ-SLP02044037.pdf","label":"IPSJ-SLP02044037"},"date":[{"dateType":"Available","dateValue":"2004-12-16"}],"format":"application/pdf","billing":["billing_file"],"filename":"IPSJ-SLP02044037.pdf","filesize":[{"value":"1.2 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":"c7c11667-814a-4f09-98b1-9602408c68c1","displaytype":"detail","licensetype":"license_note","license_note":"Copyright (c) 2002 by the Information Processing Society of Japan"}]},"item_4_creator_5":{"attribute_name":"著者名","attribute_type":"creator","attribute_value_mlt":[{"creatorNames":[{"creatorName":"西田, 晶史"},{"creatorName":"秋田, 祐哉"},{"creatorName":"河原, 達也"}],"nameIdentifiers":[{}]}]},"item_4_creator_6":{"attribute_name":"著者名(英)","attribute_type":"creator","attribute_value_mlt":[{"creatorNames":[{"creatorName":"Masafumi, Nishida","creatorNameLang":"en"},{"creatorName":"Yuya, Akita","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":"討論音声を対象とした教師なし話者インデキシングとそれを用いた音声認識について報告する．討論音声では，話者の交替が頻繁に発生し，継続時間の短い発話が多く，発話時間のばらつきが大きいため，画一的なモデルで話者インデキシングを行うのが困難である．そこで，BICに基づいて最適な話者モデル（GMMまたはVQ）を選択する方式を提案する．本方式では，発話時間の短い音声に対してVQモデル，長い音声に対してはGMMモデルが選択される枠組みを実現する．実際の討論音声に対して，従来法に比べて高いインデキシング精度を得ることができた．次に，討論音声認識のための音響・言語モデルについて検討を行う．話者インデキシング結果に基づいて音響モデルを適応することにより，音声認識制度の改善を得ることができた．","subitem_description_type":"Other"}]},"item_4_description_8":{"attribute_name":"論文抄録(英)","attribute_value_mlt":[{"subitem_description":"This paper addresses unsupervised speaker indexing for discussion audio archives. In discussions, the speaker changes frequently, thus the duration of utterances is very short and its variation is large, which causes significant problems in applying conventional methods such as model adaptation and Variance-BIC (Bayesian Information Criterion) methods. We propose a flexible framework that selects an optimal speaker model (GMM or VQ) based on the BIC according to the duration of utterances. When the speech segment is short, the simple and robust VQ-based method is expected to be chosen, while GMM will be reliably trained for long segments. For a discussion archive, it is demonstrated that the proposed method achieves higher indexing performance than that of conventional methods. The speaker index is useful for speaker adaptation of the acoustic model, which improves the performance of automatic speech recognition.","subitem_description_type":"Other"}]},"item_4_biblio_info_10":{"attribute_name":"書誌情報","attribute_value_mlt":[{"bibliographicPageEnd":"224","bibliographic_titles":[{"bibliographic_title":"情報処理学会研究報告音声言語情報処理（SLP）"}],"bibliographicPageStart":"219","bibliographicIssueDates":{"bibliographicIssueDate":"2002-12-16","bibliographicIssueDateType":"Issued"},"bibliographicIssueNumber":"121(2002-SLP-044)","bibliographicVolumeNumber":"2002"}]},"relation_version_is_last":true,"weko_creator_id":"1"},"links":{},"id":57303,"updated":"2025-01-21T15:29:42.199797+00:00"}