{"links":{},"id":234642,"metadata":{"_oai":{"id":"oai:ipsj.ixsq.nii.ac.jp:00234642","sets":["1164:5064:11558:11626"]},"path":["11626"],"owner":"44499","recid":"234642","title":["話者ダイアライゼーションを用いたマルチタスク学習によるEnd-to-End複数話者音声認識の改善"],"pubdate":{"attribute_name":"公開日","attribute_value":"2024-06-07"},"_buckets":{"deposit":"6fb8e40e-bb6d-4aa0-a6c4-791b2819023c"},"_deposit":{"id":"234642","pid":{"type":"depid","value":"234642","revision_id":0},"owners":[44499],"status":"published","created_by":44499},"item_title":"話者ダイアライゼーションを用いたマルチタスク学習によるEnd-to-End複数話者音声認識の改善","author_link":["639632","639631"],"item_titles":{"attribute_name":"タイトル","attribute_value_mlt":[{"subitem_title":"話者ダイアライゼーションを用いたマルチタスク学習によるEnd-to-End複数話者音声認識の改善"}]},"item_keyword":{"attribute_name":"キーワード","attribute_value_mlt":[{"subitem_subject":"ポスターセッション1","subitem_subject_scheme":"Other"}]},"item_type_id":"4","publish_date":"2024-06-07","item_4_text_3":{"attribute_name":"著者所属","attribute_value_mlt":[{"subitem_text_value":"奈良先端科学技術大学院大学"},{"subitem_text_value":"奈良先端科学技術大学院大学"}]},"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/234642/files/IPSJ-MUS24140030.pdf","label":"IPSJ-MUS24140030.pdf"},"date":[{"dateType":"Available","dateValue":"2026-06-07"}],"format":"application/pdf","billing":["billing_file"],"filename":"IPSJ-MUS24140030.pdf","filesize":[{"value":"928.0 kB"}],"mimetype":"application/pdf","priceinfo":[{"tax":["include_tax"],"price":"660","billingrole":"5"},{"tax":["include_tax"],"price":"330","billingrole":"6"},{"tax":["include_tax"],"price":"0","billingrole":"21"},{"tax":["include_tax"],"price":"0","billingrole":"44"}],"accessrole":"open_date","version_id":"73a5d1a3-5730-4796-85bf-8990583c6b93","displaytype":"detail","licensetype":"license_note","license_note":"Copyright (c) 2024 by the Information Processing Society of Japan"}]},"item_4_creator_5":{"attribute_name":"著者名","attribute_type":"creator","attribute_value_mlt":[{"creatorNames":[{"creatorName":"平野, 雄太"}],"nameIdentifiers":[{}]},{"creatorNames":[{"creatorName":"サクティ, サクリアニ"}],"nameIdentifiers":[{}]}]},"item_4_source_id_9":{"attribute_name":"書誌レコードID","attribute_value_mlt":[{"subitem_source_identifier":"AN10438388","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_source_id_11":{"attribute_name":"ISSN","attribute_value_mlt":[{"subitem_source_identifier":"2188-8752","subitem_source_identifier_type":"ISSN"}]},"item_4_description_7":{"attribute_name":"論文抄録","attribute_value_mlt":[{"subitem_description":"本研究では，End-to-End 複数話者音声認識モデルの認識精度改善を目的とした，話者ダイアライゼーションを用いたマルチタスク学習を提案する．エンコーダの中間層に話者ダイアライゼーションブランチを導入し，これが認識精度改善に寄与するかについて検証する．また，話者ダイアライゼーションの推定結果を考慮したエンコーダ出力を得るために，話者ダイアライゼーションの推定結果をエンコーダに入力する自己条件付けフィードバック機構を提案し，その効果についても実験的検証を行う．実験において，話者ダイアライゼーションによるマルチタスク学習と自己条件付けフィードバックを同時に用いたとき，ベースラインの認識精度からの改善が見られた．この結果から，「誰がいつ話したか」という情報が，End-to-End 複数話者音声認識モデルのエンコーダのモデリングに有効であることが示唆される．","subitem_description_type":"Other"}]},"item_4_biblio_info_10":{"attribute_name":"書誌情報","attribute_value_mlt":[{"bibliographicPageEnd":"3","bibliographic_titles":[{"bibliographic_title":"研究報告音楽情報科学（MUS）"}],"bibliographicPageStart":"1","bibliographicIssueDates":{"bibliographicIssueDate":"2024-06-07","bibliographicIssueDateType":"Issued"},"bibliographicIssueNumber":"30","bibliographicVolumeNumber":"2024-MUS-140"}]},"relation_version_is_last":true,"weko_creator_id":"44499"},"created":"2025-01-19T01:36:29.101457+00:00","updated":"2025-01-19T09:44:26.230915+00:00"}