{"id":209762,"metadata":{"_oai":{"id":"oai:ipsj.ixsq.nii.ac.jp:00209762","sets":["1164:5159:10515:10530"]},"path":["10530"],"owner":"44499","recid":"209762","title":["Comparison of End-to-End Models for Joint Speaker and Speech Recognition"],"pubdate":{"attribute_name":"公開日","attribute_value":"2021-02-24"},"_buckets":{"deposit":"9546a5e5-72ef-4e43-bcc5-cc2082721a3f"},"_deposit":{"id":"209762","pid":{"type":"depid","value":"209762","revision_id":0},"owners":[44499],"status":"published","created_by":44499},"item_title":"Comparison of End-to-End Models for Joint Speaker and Speech Recognition","author_link":["529639","529635","529634","529642","529643","529638","529636","529637","529641","529640"],"item_titles":{"attribute_name":"タイトル","attribute_value_mlt":[{"subitem_title":"Comparison of End-to-End Models for Joint Speaker and Speech Recognition"},{"subitem_title":"Comparison of End-to-End Models for Joint Speaker and Speech Recognition","subitem_title_language":"en"}]},"item_keyword":{"attribute_name":"キーワード","attribute_value_mlt":[{"subitem_subject":"SP1","subitem_subject_scheme":"Other"}]},"item_type_id":"4","publish_date":"2021-02-24","item_4_text_3":{"attribute_name":"著者所属","attribute_value_mlt":[{"subitem_text_value":"Graduate School of Informatics, Kyoto University / National Institute of Information Communications and Technology (NICT)"},{"subitem_text_value":"National Institute of Information Communications and Technology (NICT)"},{"subitem_text_value":"National Institute of Information Communications and Technology (NICT)"},{"subitem_text_value":"National Institute of Information Communications and Technology (NICT)"},{"subitem_text_value":"National Institute of Information Communications and Technology (NICT)"}]},"item_4_text_4":{"attribute_name":"著者所属(英)","attribute_value_mlt":[{"subitem_text_value":"Graduate School of Informatics, Kyoto University / National Institute of Information Communications and Technology (NICT)","subitem_text_language":"en"},{"subitem_text_value":"National Institute of Information Communications and Technology (NICT)","subitem_text_language":"en"},{"subitem_text_value":"National Institute of Information Communications and Technology (NICT)","subitem_text_language":"en"},{"subitem_text_value":"National Institute of Information Communications and Technology (NICT)","subitem_text_language":"en"},{"subitem_text_value":"National Institute of Information Communications and Technology (NICT)","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/209762/files/IPSJ-SLP21136024.pdf","label":"IPSJ-SLP21136024.pdf"},"format":"application/pdf","billing":["billing_file"],"filename":"IPSJ-SLP21136024.pdf","filesize":[{"value":"1.8 MB"}],"mimetype":"application/pdf","priceinfo":[{"tax":["include_tax"],"price":"0","billingrole":"22"},{"tax":["include_tax"],"price":"0","billingrole":"44"}],"accessrole":"open_login","version_id":"14b8f449-c5dc-4dcf-965a-6af362a998ca","displaytype":"detail","licensetype":"license_note","license_note":"Copyright (c) 2021 by the Institute of Electronics, Information and Communication Engineers This SIG report is only available to those in membership of the SIG."}]},"item_4_creator_5":{"attribute_name":"著者名","attribute_type":"creator","attribute_value_mlt":[{"creatorNames":[{"creatorName":"Kak, Soky"}],"nameIdentifiers":[{}]},{"creatorNames":[{"creatorName":"Sheng, Li"}],"nameIdentifiers":[{}]},{"creatorNames":[{"creatorName":"Masato, Mimura"}],"nameIdentifiers":[{}]},{"creatorNames":[{"creatorName":"Chenhui, Chu"}],"nameIdentifiers":[{}]},{"creatorNames":[{"creatorName":"Tatsuya, Kawahara"}],"nameIdentifiers":[{}]}]},"item_4_creator_6":{"attribute_name":"著者名(英)","attribute_type":"creator","attribute_value_mlt":[{"creatorNames":[{"creatorName":"Kak, Soky","creatorNameLang":"en"}],"nameIdentifiers":[{}]},{"creatorNames":[{"creatorName":"Sheng, Li","creatorNameLang":"en"}],"nameIdentifiers":[{}]},{"creatorNames":[{"creatorName":"Masato, Mimura","creatorNameLang":"en"}],"nameIdentifiers":[{}]},{"creatorNames":[{"creatorName":"Chenhui, Chu","creatorNameLang":"en"}],"nameIdentifiers":[{}]},{"creatorNames":[{"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_source_id_11":{"attribute_name":"ISSN","attribute_value_mlt":[{"subitem_source_identifier":"2188-8663","subitem_source_identifier_type":"ISSN"}]},"item_4_description_7":{"attribute_name":"論文抄録","attribute_value_mlt":[{"subitem_description":"In this paper, we investigate the effectiveness of using speaker information on the performance of speaker- imbalanced automatic speech recognition (ASR). We identify major speakers and combine other speakers who have a small size of speech, and make a systematic comparison of three methods that use speaker information for ASR including speaker attribute augmentation (SAug), multi-task learning (MTL), and adversarial learning (AL). We conduct experiments on a large spontaneous speech corpus of the Extraordinary Chambers in the Courts of Cambodia (ECCC) and an open Khmer text-to-speech corpus. As a result, we find that the use of speaker clustering information improves ASR performance including new speakers. Moreover, AL achieves better performance and more robustness in the speaker-independent setting compared to the other methods. It reduces errors of the baseline model by 4.32%, 5.46%, and 16.10% for the closed test, open test, and out-of-domain test, respectively.","subitem_description_type":"Other"}]},"item_4_description_8":{"attribute_name":"論文抄録(英)","attribute_value_mlt":[{"subitem_description":"In this paper, we investigate the effectiveness of using speaker information on the performance of speaker- imbalanced automatic speech recognition (ASR). We identify major speakers and combine other speakers who have a small size of speech, and make a systematic comparison of three methods that use speaker information for ASR including speaker attribute augmentation (SAug), multi-task learning (MTL), and adversarial learning (AL). We conduct experiments on a large spontaneous speech corpus of the Extraordinary Chambers in the Courts of Cambodia (ECCC) and an open Khmer text-to-speech corpus. As a result, we find that the use of speaker clustering information improves ASR performance including new speakers. Moreover, AL achieves better performance and more robustness in the speaker-independent setting compared to the other methods. It reduces errors of the baseline model by 4.32%, 5.46%, and 16.10% for the closed test, open test, and out-of-domain test, respectively.","subitem_description_type":"Other"}]},"item_4_biblio_info_10":{"attribute_name":"書誌情報","attribute_value_mlt":[{"bibliographicPageEnd":"5","bibliographic_titles":[{"bibliographic_title":"研究報告音声言語情報処理（SLP）"}],"bibliographicPageStart":"1","bibliographicIssueDates":{"bibliographicIssueDate":"2021-02-24","bibliographicIssueDateType":"Issued"},"bibliographicIssueNumber":"24","bibliographicVolumeNumber":"2021-SLP-136"}]},"relation_version_is_last":true,"weko_creator_id":"44499"},"updated":"2025-01-19T18:23:58.121562+00:00","created":"2025-01-19T01:11:03.039806+00:00","links":{}}