asEars: Single-Sided Deafness Assistive Glasses

Core Collaborators: Etsushi Nozaki*, Tomomi Kanai*, Koki Shinjo*, Yuki Ichikawa**, Kotatsu Shiraki*, Haruka Kyuko†, Akinori Kashio‡, Yoshihiro Kawahara*

*The University of Tokyo  **Chiba Institute of Technology  †Tokyo University of the Arts   ‡The University of Tokyo Hospital

Background and Goal

Single-sided deafness (SSD) significantly restricts social participation in hearing/speaking cultures due to the person’s difficulty hearing conversations on their deaf side. Although hearing aids for SSD are effective in social situations, the acceptance rate remains low at 4% among SSD patients. Because I also have single-sided deafness, I found it challenging to join a conversation in noisy situations like parties, and I couldn’t enjoy it so much. To address these problems, our goal is to design and develop an assistive device with better effectiveness than existing hearing aids for SSD and a higher level of wearability, appearance, and social acceptance. 

Approach

In this research, we began with a background study of 51 people with single-sided deafness (SSD) to pinpoint challenges with SSD and limitations of existing hearing aids. Guided by those insights, our team—including two members with SSD—through three prototype iterations of an assistive device, asEars (shown above), designed for real-world acceptability. I lead a team with five core members and 20+ collaborators across the consumer electronics, eyewear, and hearing-aid industries, as well as otolaryngology. The resulting system delivers 6 ms end-to-end latency; cartilage / bone conduction; beamforming optimized for SSD; power-managed firmware giving ~10 h battery life.” for everyday use.

To evaluate the proposed device in user experience aspects and clinical aspects, we conducted a comparative study with ten individuals with SSD, where they were asked to use the proposed device and traditional hearing aids for SSD (CROS hearing aids) for two weeks each. According to their daily questionnaire, interview results, and hearing-in-noise test results, we showed that participants prefer our device over CROS hearing aids, while being effective in terms of hearing in noise.

Ongoing Work

We plan to further increase the effectiveness and acceptability of our device by incorporating deep-learning-based signal processing with computing assistance from the cloud. The research is currently funded by a national grant in Japan: hearing aids that work with deep learning on the cloud, focusing on the nature of conversation and supported by clinical collaborations with the University of Tokyo Otolaryngology.

Impact

• Honorable Mention Award 🥇 at ACM CHI 2023 (Top 5%)

• Interdisciplinary Telecommunications Student Research Award 🏆 (2nd Prize in Japan)

• Best Poster Award (Top 5%), Dept. of EEIS, The University of Tokyo

Not only recognized in academia, but also generated strong public impact: 100+ trial requests from people with single-sided deafness, frequent discussion in SSD advocacy groups, and extensive media coverage.

Acknowledgements

• This work was supported by JSPS Grant-in-Aid for JSPS Fellows Grant Number JP21J22274, AMED under Grant Number JP21dk0310104, JST ERATO Grant Number JPMJER1501, and Value Exchange Engineering, a joint research project between R4D, Mercari Inc., and the RIISE.

• We would also like to thank Hironobu Kasashima, Satoshi Nakamura, Rei Hosoi, Koki Shinjo, Masahiko Horiuchi, Takaaki Umada, Toshinari Shimokawa, Keisuke Shiro, Seiji Nakagawa, and Tomy Kamada for design advice on asEars, Katsufumi Matsui, Takeshi Naemura, Keiichi Takaki, Takatoshi Yoshida, Fumikazu Saze, and Koya Narumi for the advice of the research design, Jie Qi and Andrew bunnie Huang for directing the Research at Scale course, Makichie Co., Ltd. for providing CROS hearing aids, Ryohei Harada, Akie Kochiya, Kenji Ito, and Yui Takaki for helping with the experiments.

Links

Research Publications

• asEars: Designing and Evaluating the User Experience of Wearable Assistive Devices for Single-Sided Deafness, CHI2023🥇 (first author)

• A Randomized Crossover Study in Single-Sided Deafness Comparing a Cartilage Conduction CROS System and an Air-Conduction CROS System, Otology & Neurotology 45 (6) (first author)

• Higher SNR or Lower Latency? Investigating the Effects of SNR and Latency on Hearing in Noise with Single-sided Deafness, UIST 2025 (first author)

• Micro-Gesture Recognition of Tongue via Bone Conduction Sound, UIST 2024 (2nd author)

• Toward Continuous Finger Positioning on Ear Using Bone Conduction Speaker, SenSys 2022 (2nd author)

Press Coverage

• Aljazeera, "The Stream: Rethinking disability and assistive technology," 2018/3/15.

• Medgadget, "asEars Assists People With Single-Sided Deafness – With Style," 2018/4/16.

• Wareable, "These stylish glasses will help people with single-sided deafness," 2018/3/13.

• テレビ東京, "生きるを伝える：片耳難聴を支援するメガネ型デバイスの開発者の裏側, " 2024/08/10.

• テレビ朝日, "発進！ミライクリエイター, " 2023/8/16.

• NHK Eテレ, "クイズ#ふつうアップデート スゴい研究SP(1), " 2023/3/10．

• 日本テレビ, "nextクリエイターズ クリエイターFILE #152 聞くのが楽しい眼鏡「asEars」, " 2020/8/27．

• NHK Eテレ, "ろうを生きる 難聴を生きる “片耳難聴”当事者グループ「きこいろ」," 2020/3/28.

• 日経TechLiveX, "変わるイマドキ東大生のキャリア観 「東大→起業」という生き方（後編：現役東大生＆OB起業家大集合！）," 2018/4/27.

• 東京大学工学部, "電気系工学専攻 高木 健 さん（D3）らの研究チームの論文が、CHI 2023においてHonorable Mention Awardを受賞されました”,2023/5/30

• 東京大学高大接続研究開発センター, "キミの東大「東京大学総長賞」令和２年度 受賞者の声（学業編）―東大を目指すキミへのメッセージ”,2021/4/19

• 東京大学大学院情報理工学系研究科, "特別対談企画"研究の王道"に縛られない――大切なのはその視点を持ち続け、ニーズを察知していくこと", 2021/4/13

• Autodesk Redshift, "asEars: 片耳難聴者を対象にデザインされたユニークなメガネ型デバイス," 2018/12/4.

• 丸紅情報システムズ株式会社，"３Dプリンター導入事例 asEars様　半年間で片耳難聴者向けの補聴器を開発　最先端の道具を手に、大学生5人が挑む", 2018/11

• Elephantech, "【インタビュー】メガネ型の片耳難聴者用補聴器「asEars」プロジェクトのリーダー 高木 健さん", 2018/8/2

• Bouncy, "東大発のアイデアをSXSWへ送り出す「Todai To Texas」、2018年の注目ブースは？," 2018/3/22.

• 日経産業新聞　片耳難聴者を「眼鏡」で支援　2022/4/6

• 第23回「世界vision lab」～片耳難聴を解決するウェアラブル・デバイスを開発した「asEars」のプロジェクトリーダー、東京大学大学院工学系研究科に在籍中の高木健さんがゲスト　2022/3/6