{"metadata":{"_oai":{"id":"oai:ipsj.ixsq.nii.ac.jp:00205216","sets":["6504:10247:10254"]},"path":["10254"],"owner":"6748","recid":"205216","title":["話者・音素特徴に基づくマルチチャネル音声分離"],"pubdate":{"attribute_name":"公開日","attribute_value":"2020-02-20"},"_buckets":{"deposit":"4b30a8c6-9821-44d9-9c87-dbd59e015858"},"_deposit":{"id":"205216","pid":{"type":"depid","value":"205216","revision_id":0},"owners":[6748],"status":"published","created_by":6748},"item_title":"話者・音素特徴に基づくマルチチャネル音声分離","author_link":["509089","509092","509090","509093","509091"],"item_titles":{"attribute_name":"タイトル","attribute_value_mlt":[{"subitem_title":"話者・音素特徴に基づくマルチチャネル音声分離"}]},"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":"2020-02-20","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 / 理研"},{"subitem_text_value":"産総研"},{"subitem_text_value":"理研"},{"subitem_text_value":"京大 / 理研"}]},"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/205216/files/IPSJ-Z82-5Q-03.pdf","label":"IPSJ-Z82-5Q-03.pdf"},"date":[{"dateType":"Available","dateValue":"2020-06-19"}],"format":"application/pdf","filename":"IPSJ-Z82-5Q-03.pdf","filesize":[{"value":"1.0 MB"}],"mimetype":"application/pdf","accessrole":"open_date","version_id":"72e752ed-338c-4881-a13a-2da5c73daaa0","displaytype":"detail","licensetype":"license_note","license_note":"Copyright (c) 2020 by the Information Processing Society of Japan"}]},"item_22_creator_5":{"attribute_name":"著者名","attribute_type":"creator","attribute_value_mlt":[{"creatorNames":[{"creatorName":"Yicheng, Du"}],"nameIdentifiers":[{}]},{"creatorNames":[{"creatorName":"關口, 航平"}],"nameIdentifiers":[{}]},{"creatorNames":[{"creatorName":"坂東, 宜昭"}],"nameIdentifiers":[{}]},{"creatorNames":[{"creatorName":"Aditya, Arie Nugraha"}],"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":"代表的なブラインド音源分離手法であるマルチチャネル非負値行列因子分解 (MNMF) では，音源モデルと空間モデルが重要な役割を果たしている．最近、NMFに基づく低ランク音源モデルの代わりに、DNNに基づく深層音源モデルを用いた半教師あり音源分離手法が提案されている．本研究では，話者特徴と音素特徴の二種類の潜在変数を内包する深層音源モデルを定式化し，各空間内での話者や音素の分離度が高くなるような学習法を提案する。任意話者の混合音を用いた実験により，提案手法の有用性を検証する．","subitem_description_type":"Other"}]},"item_22_biblio_info_10":{"attribute_name":"書誌情報","attribute_value_mlt":[{"bibliographicPageEnd":"194","bibliographic_titles":[{"bibliographic_title":"第82回全国大会講演論文集"}],"bibliographicPageStart":"193","bibliographicIssueDates":{"bibliographicIssueDate":"2020-02-20","bibliographicIssueDateType":"Issued"},"bibliographicIssueNumber":"1","bibliographicVolumeNumber":"2020"}]},"relation_version_is_last":true,"weko_creator_id":"6748"},"id":205216,"updated":"2025-01-19T19:52:13.367837+00:00","links":{},"created":"2025-01-19T01:07:23.073126+00:00"}