{"metadata":{"_oai":{"id":"oai:ipsj.ixsq.nii.ac.jp:00214920","sets":["6504:10735:10808"]},"path":["10808"],"owner":"44499","recid":"214920","title":["逐次的音源検出と高速マルチチャネル非負値因子分解に基づくオンライン音源分離"],"pubdate":{"attribute_name":"公開日","attribute_value":"2021-03-04"},"_buckets":{"deposit":"b834d977-0179-4154-ac2f-45ec41d49694"},"_deposit":{"id":"214920","pid":{"type":"depid","value":"214920","revision_id":0},"owners":[44499],"status":"published","created_by":44499},"item_title":"逐次的音源検出と高速マルチチャネル非負値因子分解に基づくオンライン音源分離","author_link":["553130","553129","553127","553128"],"item_titles":{"attribute_name":"タイトル","attribute_value_mlt":[{"subitem_title":"逐次的音源検出と高速マルチチャネル非負値因子分解に基づくオンライン音源分離"}]},"item_keyword":{"attribute_name":"キーワード","attribute_value_mlt":[{"subitem_subject":"人工知能と認知科学","subitem_subject_scheme":"Other"}]},"item_type_id":"22","publish_date":"2021-03-04","item_language":{"attribute_name":"言語","attribute_value_mlt":[{"subitem_language":"jpn"}]},"item_22_text_3":{"attribute_name":"著者所属","attribute_value_mlt":[{"subitem_text_value":"京大"},{"subitem_text_value":"京大"},{"subitem_text_value":"京大"},{"subitem_text_value":"京大"}]},"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/214920/files/IPSJ-Z83-7N-05.pdf","label":"IPSJ-Z83-7N-05.pdf"},"date":[{"dateType":"Available","dateValue":"2021-12-28"}],"format":"application/pdf","filename":"IPSJ-Z83-7N-05.pdf","filesize":[{"value":"358.4 kB"}],"mimetype":"application/pdf","accessrole":"open_date","version_id":"5f24ef77-deb2-4760-a21c-8c83c57fb4b5","displaytype":"detail","licensetype":"license_note","license_note":"Copyright (c) 2021 by the Information Processing Society of Japan"}]},"item_22_creator_5":{"attribute_name":"著者名","attribute_type":"creator","attribute_value_mlt":[{"creatorNames":[{"creatorName":"藤原, 啓梧"}],"nameIdentifiers":[{}]},{"creatorNames":[{"creatorName":"関口, 航平"}],"nameIdentifiers":[{}]},{"creatorNames":[{"creatorName":"Yicheng, Du"}],"nameIdentifiers":[{}]},{"creatorNames":[{"creatorName":"吉井, 和佳"}],"nameIdentifiers":[{}]}]},"item_resource_type":{"attribute_name":"資源タイプ","attribute_value_mlt":[{"resourceuri":"http://purl.org/coar/resource_type/c_5794","resourcetype":"conference paper"}]},"item_22_source_id_9":{"attribute_name":"書誌レコードID","attribute_value_mlt":[{"subitem_source_identifier":"AN00349328","subitem_source_identifier_type":"NCID"}]},"item_22_description_7":{"attribute_name":"論文抄録","attribute_value_mlt":[{"subitem_description":"本稿では、新規音源がたびたび現れる環境下でのオンラインマルチチャネル音源分離について述べる。最新のBSS手法である高速マルチチャネル非負値因子分解 (FastMNMF) は、分離精度に優れ、計算量が小さい点で、入力音響信号のミニバッチ分割に基づくパラメータの逐次更新に適している。しかし、音源が新たに出現した際には、パラメータ全体が大きく更新され、分離が不安定になりやすい。この問題を解決するため、本研究では、MUSIC法による音源検出に基づくFastMNMFの安定的なオンライン実行を試みる。具体的には、新規音源が見つかった場合、パラメータの一部を音源方向のステアリングベクトルで初期化し、反復音源ステアリング法または反復射影法の部分適応により、新規音源に関するパラメータの更新を重点的に行ったのち、全パラメータの更新を行う。実験により提案手法を評価する。","subitem_description_type":"Other"}]},"item_22_biblio_info_10":{"attribute_name":"書誌情報","attribute_value_mlt":[{"bibliographicPageEnd":"234","bibliographic_titles":[{"bibliographic_title":"第83回全国大会講演論文集"}],"bibliographicPageStart":"233","bibliographicIssueDates":{"bibliographicIssueDate":"2021-03-04","bibliographicIssueDateType":"Issued"},"bibliographicIssueNumber":"1","bibliographicVolumeNumber":"2021"}]},"relation_version_is_last":true,"weko_creator_id":"44499"},"id":214920,"updated":"2025-01-19T16:24:23.029224+00:00","links":{},"created":"2025-01-19T01:15:42.661919+00:00"}