{"links":{},"id":2009886,"metadata":{"_oai":{"id":"oai:ipsj.ixsq.nii.ac.jp:02009886","sets":["1164:5159:1771204180543:1779319617078"]},"path":["1779319617078"],"owner":"80578","recid":"2009886","title":["口唇映像を用いたマルチモーダル音声認識における構音障害者のための話者適応の検討"],"pubdate":{"attribute_name":"PubDate","attribute_value":"2026-05-29"},"_buckets":{"deposit":"eebd4663-b885-4fc9-913d-2ab102bf00a5"},"_deposit":{"id":"2009886","pid":{"type":"depid","value":"2009886","revision_id":0},"owners":[80578],"status":"published","created_by":80578},"item_title":"口唇映像を用いたマルチモーダル音声認識における構音障害者のための話者適応の検討","author_link":[],"item_titles":{"attribute_name":"タイトル","attribute_value_mlt":[{"subitem_title":"口唇映像を用いたマルチモーダル音声認識における構音障害者のための話者適応の検討","subitem_title_language":"ja"},{"subitem_title":"Speaker Adaptation for Dysarthric Speakers in Multimodal Speech Recognition Using Lip Movements","subitem_title_language":"en"}]},"item_keyword":{"attribute_name":"キーワード","attribute_value_mlt":[{"subitem_subject":"SLP","subitem_subject_scheme":"Other"}]},"item_type_id":"4","publish_date":"2026-05-29","item_4_text_3":{"attribute_name":"著者所属","attribute_value_mlt":[{"subitem_text_value":"岡山大学"},{"subitem_text_value":"岡山大学"}]},"item_4_text_4":{"attribute_name":"著者所属(英)","attribute_value_mlt":[{"subitem_text_value":"Okayama University","subitem_text_language":"en"},{"subitem_text_value":"Okayama University","subitem_text_language":"en"}]},"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/2009886/files/IPSJ-SLP26160012.pdf","label":"IPSJ-SLP26160012.pdf"},"date":[{"dateType":"Available","dateValue":"2028-05-29"}],"format":"application/pdf","billing":["billing_file"],"filename":"IPSJ-SLP26160012.pdf","filesize":[{"value":"1.2 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":"23174189-0765-4e68-aff4-2429c8d674f1","displaytype":"detail","licensetype":"license_note","license_note":"Copyright (c) 2026 by the Information Processing Society of Japan"}]},"item_4_creator_5":{"attribute_name":"著者名","attribute_type":"creator","attribute_value_mlt":[{"creatorNames":[{"creatorName":"木山,大輔"}]},{"creatorNames":[{"creatorName":"原,直"}]}]},"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":"本研究では，構音障害者の会話支援を目的として，日本語における特定話者の音声認識精度の向上を目指す．構音障害者の発声は健常者とは大きく異なり，発話が不明瞭となる傾向があるため，従来の音声認識技術では高精度な認識が困難である．そこで，口唇映像と音声の両モダリティを利用した視聴覚音声認識（Audio-Visual Speech Recognition：AVSR）に着目し，AV-HuBERTを基盤とした段階的ファインチューニング手法を提案する．さらに，ターゲット話者の学習データ量を変化させ，データによる話者適応の効果を検証する．本手法では，まず健常話者の日本語データを用いた言語適応，次に複数の構音障害話者による話者非依存適応，最後にターゲット話者による話者依存適応を行う．評価指標には音素誤り率（Phoneme Error Rate: PER）を用いる．実験の結果，AV-HuBERTが日本語への適応に成功していることが確認された．また，構音障害話者への適応評価では，提案方式が最も低いPERを示した．さらに，ターゲット話者の学習データは約50文で大きな性能改善が得られ，それ以上の増加による効果は限定的であった．これにより，少量データでの話者適応の有効性が示された．","subitem_description_type":"Other"}]},"item_4_biblio_info_10":{"attribute_name":"書誌情報","attribute_value_mlt":[{"bibliographicPageEnd":"7","bibliographic_titles":[{"bibliographic_title":"研究報告音声言語情報処理（SLP）"}],"bibliographicPageStart":"1","bibliographicIssueDates":{"bibliographicIssueDate":"2026-05-29","bibliographicIssueDateType":"Issued"},"bibliographicIssueNumber":"12","bibliographicVolumeNumber":"2026-SLP-160"}]},"relation_version_is_last":true,"weko_creator_id":"80578"},"created":"2026-05-21T07:32:38.151493+00:00","updated":"2026-05-21T07:32:42.087483+00:00"}