{"created":"2025-01-18T23:19:57.375462+00:00","metadata":{"_oai":{"id":"oai:ipsj.ixsq.nii.ac.jp:00056564","sets":["1164:5159:5160:5161"]},"path":["5161"],"owner":"1","recid":"56564","title":["オンライン変分ベイズ学習に基づくモデル比較を用いた音声区間検出"],"pubdate":{"attribute_name":"公開日","attribute_value":"2009-01-30"},"_buckets":{"deposit":"9af9e719-d6ba-406e-9b10-ce20cf2f501e"},"_deposit":{"id":"56564","pid":{"type":"depid","value":"56564","revision_id":0},"owners":[1],"status":"published","created_by":1},"item_title":"オンライン変分ベイズ学習に基づくモデル比較を用いた音声区間検出","author_link":["0","0"],"item_titles":{"attribute_name":"タイトル","attribute_value_mlt":[{"subitem_title":"オンライン変分ベイズ学習に基づくモデル比較を用いた音声区間検出"},{"subitem_title":"Using Online Model Comparison in the Variational Bayes Framework : an Application to Voice Activity Detection","subitem_title_language":"en"}]},"item_type_id":"4","publish_date":"2009-01-30","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":"NTTコミュニケーション基礎科学研究所"},{"subitem_text_value":"NTT コミュニケーション基礎科学研究所"},{"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":"School of Informatics, Kyoto University","subitem_text_language":"en"},{"subitem_text_value":"NTT Communication Science Laboratories","subitem_text_language":"en"},{"subitem_text_value":"NTT Communication Science Laboratories","subitem_text_language":"en"},{"subitem_text_value":"School of Informatics, Kyoto University","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/56564/files/IPSJ-SLP09075003.pdf","label":"IPSJ-SLP09075003"},"date":[{"dateType":"Available","dateValue":"2011-01-30"}],"format":"application/pdf","billing":["billing_file"],"filename":"IPSJ-SLP09075003.pdf","filesize":[{"value":"1.3 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":"85022252-42cf-4ab9-8ed6-e2b4174cfed2","displaytype":"detail","licensetype":"license_note","license_note":"Copyright (c) 2009 by the Information Processing Society of Japan"}]},"item_4_creator_5":{"attribute_name":"著者名","attribute_type":"creator","attribute_value_mlt":[{"creatorNames":[{"creatorName":"クーナポ, ダビド"},{"creatorName":"渡部, 晋治"},{"creatorName":"中村, 篤"},{"creatorName":"河原, 達也"}],"nameIdentifiers":[{}]}]},"item_4_creator_6":{"attribute_name":"著者名(英)","attribute_type":"creator","attribute_value_mlt":[{"creatorNames":[{"creatorName":"David, Cournapeau","creatorNameLang":"en"},{"creatorName":"Shinji, Watanabe","creatorNameLang":"en"},{"creatorName":"Atsushi, Nakamura","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":"教師なし・オンラインの音声区間検出 (VAD) 方法を提案する。オンライン EM は学習データのない未知の環境にも適用できる枠組みであるが、雑音のみの区間や音声のみの区間が連続すると、モデルの更新が適切に行われないという問題があった。これに対して、提案手法は変分ベイズ EM(VB-EM) 学習に基づいており、その過程で得られる自由エネルギー (Free Energy) をモデルの信頼度比較に利用するものである。VB-EM をオンライン学習に定式化し 、モでルパラメータとモデル信頼度の推定を同時・逐次的に行う。CENSREC-1-Cを用いた音声区間検出の評価実験により、提案手法が従来のオンライン EM よりも有意に効果的であることを確認した。","subitem_description_type":"Other"}]},"item_4_description_8":{"attribute_name":"論文抄録(英)","attribute_value_mlt":[{"subitem_description":"We propose an unsupervised online method for Voice Activity Detection (VAD). The online EM can be applied to any new environments with no training data, but falls in unreliable estimations when the noise-only or speech-only segments last for a long time. The proposed method is based on the Variational Bayes (VB) approach to EM algorithm, and uses Free Energy, which is computed during the estimation process, to assess the model reliability in parallel. An online variation of the VB-EM is formulated for sequential estimation of both model parameters and model comparison measure. An experimental evaluation using the CENSREC-1-C database demonstrates that the proposed method significantly outperformed the conventional online EM method.","subitem_description_type":"Other"}]},"item_4_biblio_info_10":{"attribute_name":"書誌情報","attribute_value_mlt":[{"bibliographicPageEnd":"18","bibliographic_titles":[{"bibliographic_title":"情報処理学会研究報告音声言語情報処理（SLP）"}],"bibliographicPageStart":"13","bibliographicIssueDates":{"bibliographicIssueDate":"2009-01-30","bibliographicIssueDateType":"Issued"},"bibliographicIssueNumber":"10(2009-SLP-075)","bibliographicVolumeNumber":"2009"}]},"relation_version_is_last":true,"weko_creator_id":"1"},"id":56564,"updated":"2025-01-21T15:29:50.025491+00:00","links":{}}