At the international conference UbiComp 2025 (ACM International Joint Conference on Pervasive and Ubiquitous Computing), held on October 14–16 at Aalto University in Espoo, Finland, the research group including Researcher Xuefu Dong, Associate Professor Yuuki Nishiyama, and Professor Kaoru Sezaki was selected for the Best Poster Award.
The award-winning study, “Recognizing Hidden-in-the-Ear Private Key for Reliable Silent Speech Interface Using Multi-Task Learning,” explored a silent speech pathway that learns a person-specific “hidden-in-the-ear” key by jointly modeling whisper-like audio and subtle ear-canal and muscle movements, and demonstrated a method to enhance authentication and reliability in earable-based silent speech interfaces.
Authors: Xuefu Dong, Liqiang Xu, Lixing He, Zengyi Han, Kenneth Christofferson, Yifei Chen, Akihito Taya, Yuuki Nishiyama, Kaoru Sezaki
Title: Recognizing Hidden-in-the-Ear Private Key for Reliable Silent Speech Interface Using Multi-Task Learning
🔗UbiComp 2025：https://www.ubicomp.org/ubicomp-iswc-2025/
🔗Follow our network: https://linktr.ee/CSISut
10月14日～16日にフィンランド・エスポーのアールト大学（Aalto University）で開催された国際会議 UbiComp 2025（ACM International Joint Conference on Pervasive and Ubiquitous Computing） にて、董学甫 研究員、西山勇毅 准教授、瀬崎薫 教授らの研究がベストポスター賞に選ばれました。
受賞した研究「Recognizing Hidden-in-the-Ear Private Key for Reliable Silent Speech Interface Using Multi-Task Learning」では、ささやき声のような音と耳道や筋肉の微細な動きを同時にモデル化することで、個人固有の「耳に隠された鍵」を学習しました。これにより、イヤラブルデバイスを用いたサイレントスピーチインターフェースにおける認証と信頼性を高める手法を示しました。
発表者: Xuefu Dong, Liqiang Xu, Lixing He, Zengyi Han, Kenneth Christofferson, Yifei Chen, Akihito Taya, Yuuki Nishiyama, Kaoru Sezaki
タイトル: Recognizing Hidden-in-the-Ear Private Key for Reliable Silent Speech Interface Using Multi-Task Learning
🔗UbiComp 2025：https://www.ubicomp.org/ubicomp-iswc-2025/（英語）
🔗CSIS公式SNS: https://linktr.ee/CSISut
授賞式の様子 / At the award ceremony
Best Poster Awardの賞状 / Best Poster Award certificate