pith. sign in

arxiv: 2509.20799 · v5 · submitted 2025-09-25 · 💻 cs.HC · cs.SD

AuthGlass: Benchmarking Voice Liveness Detection and Authentication on Smart Glasses via Comprehensive Acoustic Features

Weiye Xu , Zhang Jiang , Siqi Zheng , Xiyuxing Zhang , Changhao Zhang , Jian Liu , Weiqiang Wang , Yuntao Wang This is my paper

Pith reviewed 2026-05-18 14:36 UTC · model grok-4.3

classification 💻 cs.HC cs.SD
keywords voice liveness detectionsmart glassesvoice authenticationacoustic featuresspoofing attacksmulti-channel audioAuthGlass dataset
0
0 comments X p. Extension

The pith

Smart glasses can use multi-channel audio to detect live voices and authenticate users more accurately than prior methods.

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The paper collects a new dataset of 16-channel audio recordings from 42 subjects together with two categories of spoofing attacks to study voice security on smart glasses. It introduces AuthG-Live, a method that analyzes sound field properties to separate real speech from fakes, and AuthG-Net, a model that fuses features across acoustic modalities for user authentication. Benchmark tests against seven liveness detectors and four authentication systems show the new approaches reach top results on four tasks while remaining effective under varied recording conditions. Releasing the full dataset lets others develop and compare secure voice systems for wearable devices.

Core claim

By building a dedicated multi-acoustic-modal dataset and designing sound-field-based liveness detection plus multi-modal fusion authentication, the work demonstrates state-of-the-art performance on voice liveness and user authentication tasks for smart glasses, with ablation studies confirming robustness across real-world usage constraints.

What carries the argument

AuthG-Live, a sound-field-based voice liveness detection method, and AuthG-Net, a multi-acoustic-modal authentication model that processes 16-channel recordings.

If this is right

  • Voice commands on smart glasses become harder to spoof with replay or synthesis attacks when sound-field and multi-channel cues are used.
  • Authentication performance improves by combining signals from all 16 microphones rather than relying on single-channel input.
  • The benchmark tasks establish a standard for comparing new liveness and authentication techniques on wearable audio devices.
  • Ablation results indicate the methods tolerate changes in subject, environment, and attack style within the collected data.

Where Pith is reading between the lines

These are editorial extensions of the paper, not claims the author makes directly.

  • The same multi-channel feature approach could be adapted to other wearables such as earbuds that also have several microphones.
  • Adding motion or proximity sensors might address edge cases where acoustic cues alone are ambiguous.
  • Device makers could embed these detection steps directly in firmware to block unauthorized voice access.

Load-bearing premise

Recordings from 42 subjects and two attack categories are representative enough of real smart-glasses use and spoofing threats that results on this dataset will hold on deployed devices.

What would settle it

Evaluating the models on audio collected from a fresh group of users wearing actual smart glasses in uncontrolled daily environments and measuring whether accuracy falls below the reported benchmark levels.

Figures

Figures reproduced from arXiv: 2509.20799 by Changhao Zhang, Jian Liu, Siqi Zheng, Weiqiang Wang, Weiye Xu, Xiyuxing Zhang, Yuntao Wang, Zhang Jiang.

Figure 1
Figure 1. Figure 1: (a) Structure of vocal tract, including vocal cords, tongue and other articulatory organs (b) Previous research revealed that [PITH_FULL_IMAGE:figures/full_fig_p005_1.png] view at source ↗
Figure 2
Figure 2. Figure 2: a) Structural and hardware demonstration of the prototype. The prototype includes 14 channels of air-conductive microphones [PITH_FULL_IMAGE:figures/full_fig_p006_2.png] view at source ↗
Figure 3
Figure 3. Figure 3: (a) The environment setting of the voice attack. i and ii are smart glasses prototype, iii is the torso–mouth simulator and iv is [PITH_FULL_IMAGE:figures/full_fig_p008_3.png] view at source ↗
Figure 4
Figure 4. Figure 4: In the restricted microphone experiment. (a) shows the three microphones selected. (b) shows the five microphones selected.(c) [PITH_FULL_IMAGE:figures/full_fig_p009_4.png] view at source ↗
Figure 4
Figure 4. Figure 4: In these cases, the microphone number was restricted and sound field feature could only be partially captured. [PITH_FULL_IMAGE:figures/full_fig_p010_4.png] view at source ↗
read the original abstract

With the rapid advancement of smart glasses, voice interaction has been widely adopted due to its naturalness and convenience. However, its practical deployment is often undermined by vulnerability to spoofing attacks, while no public dataset currently exists for voice liveness detection and authentication in smart-glasses scenarios. To address this challenge, we first collect a multi-acoustic-modal dataset comprising 16-channel audio data from 42 subjects, along with corresponding attack samples covering two attack categories. Based on insights derived from this collected data, we propose AuthG-Live, a sound-field-based voice liveness detection method, and AuthG-Net, a multi-acoustic-modal authentication model. We further benchmark seven voice liveness detection methods and four authentication methods across diverse acoustic modalities. The results demonstrate that our proposed approach achieves state-of-the-art performance on four benchmark tasks, and extensive ablation studies validate the generalizability of our methods \red{under real-world constraints}. Finally, we release this dataset, termed AuthGlass, to facilitate future research on voice liveness detection and authentication for smart glasses.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit. Tearing a paper down is the easy half of reading it; the pith above is the substance, this is the friction.

Referee Report

2 major / 1 minor

Summary. The paper collects a new multi-acoustic-modal dataset called AuthGlass comprising 16-channel audio recordings from 42 subjects together with attack samples from two categories. It introduces AuthG-Live, a sound-field-based voice liveness detection method, and AuthG-Net, a multi-acoustic-modal authentication model. The authors benchmark seven existing liveness detection methods and four authentication methods, claiming state-of-the-art results on four tasks, and present ablation studies that are said to validate generalizability under real-world constraints. The dataset is released publicly.

Significance. If the performance claims are substantiated, the work would be significant by supplying the first public dataset and tailored methods for voice security on smart glasses, an increasingly common platform. The public release of the dataset is a clear strength that directly supports reproducibility and follow-on research.

major comments (2)
  1. [Dataset Collection] Dataset Collection section: The central generalizability claims rest on the dataset of 42 subjects and two attack categories being representative of real-world voices, acoustic environments, glass form factors, and spoofing vectors; this modest scale for voice biometrics leaves open whether performance will hold for unseen synthesis methods, replay distances, or noise profiles outside the collection protocol.
  2. [Results and Evaluation] Results and Evaluation sections: The abstract and main results report SOTA performance together with ablation studies, yet the absence of full experimental details, baseline implementations, exact metrics, and error bars makes it impossible to verify whether the reported performance numbers are supported by the underlying data.
minor comments (1)
  1. [Abstract] The phrase 'under real-world constraints' in the abstract and conclusion would benefit from an explicit list of the tested constraints (e.g., specific noise levels or distances) in the main text.

Simulated Author's Rebuttal

2 responses · 1 unresolved

We thank the referee for the constructive and detailed feedback on our manuscript. We have carefully considered each major comment and provide point-by-point responses below. Where appropriate, we will revise the manuscript to address the concerns raised while maintaining the integrity of our contributions.

read point-by-point responses

  1. Referee: [Dataset Collection] Dataset Collection section: The central generalizability claims rest on the dataset of 42 subjects and two attack categories being representative of real-world voices, acoustic environments, glass form factors, and spoofing vectors; this modest scale for voice biometrics leaves open whether performance will hold for unseen synthesis methods, replay distances, or noise profiles outside the collection protocol.

    Authors: We acknowledge that 42 subjects represents a modest scale relative to some large-scale voice biometrics corpora. However, AuthGlass is the first public multi-acoustic-modal dataset collected specifically for smart glasses, involving synchronized 16-channel recordings and two distinct attack categories under controlled yet varied conditions. Subject recruitment covered diverse demographics, and the protocol incorporated multiple acoustic environments and glass form factors. In the revised manuscript, we will expand the Dataset Collection section with additional statistics on subject diversity, environmental variations, and attack simulation details. We will also add a dedicated Limitations subsection that explicitly discusses the boundaries of claimed generalizability and outlines directions for future dataset expansion. revision: partial

  2. Referee: [Results and Evaluation] Results and Evaluation sections: The abstract and main results report SOTA performance together with ablation studies, yet the absence of full experimental details, baseline implementations, exact metrics, and error bars makes it impossible to verify whether the reported performance numbers are supported by the underlying data.

    Authors: We agree that the current version lacks sufficient experimental transparency. In the revision, we will substantially expand the Results and Evaluation sections to include complete hyperparameter settings, precise descriptions of all baseline implementations (including any adaptations from original works), exact metric definitions and computation procedures, and error bars or standard deviations computed across repeated trials. We will also release the full evaluation code and scripts alongside the existing dataset release to enable independent reproduction and verification of all reported results. revision: yes

standing simulated objections not resolved
  • Whether the reported performance will hold for synthesis methods, replay distances, or noise profiles entirely outside the two attack categories and collection protocol used in the current dataset.

Circularity Check

0 steps flagged

Empirical benchmarking on newly collected dataset exhibits no circularity

full rationale

The paper collects a fresh 16-channel dataset from 42 subjects plus two attack categories, derives insights to propose AuthG-Live and AuthG-Net, then benchmarks seven liveness and four authentication methods. All reported performance numbers and ablation results are direct empirical measurements on the held-out portions of this new dataset; no equations, fitted parameters, or predictions are shown to reduce by construction to quantities derived from the same test set. No self-citation chains or uniqueness theorems are invoked to justify the central claims. The work is therefore self-contained against external benchmarks and receives the default non-finding score.

Axiom & Free-Parameter Ledger

0 free parameters · 1 axioms · 0 invented entities

The central claims rest on the assumption that the collected 42-subject, 16-channel dataset plus two attack types capture the relevant acoustic distinctions; no free parameters, axioms, or invented entities are explicitly listed in the abstract.

axioms (1)
  • domain assumption The multi-acoustic-modal recordings from 42 subjects and the chosen attack samples are representative of real-world smart-glasses conditions.
    Stated as the basis for proposing AuthG-Live and AuthG-Net and for claiming generalizability under real-world constraints.

pith-pipeline@v0.9.0 · 5742 in / 1241 out tokens · 46411 ms · 2026-05-18T14:36:59.138677+00:00 · methodology

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Lean theorems connected to this paper

Citations machine-checked in the Pith Canon. Every link opens the source theorem in the public Lean library.

What do these tags mean?
matches
The paper's claim is directly supported by a theorem in the formal canon.
supports
The theorem supports part of the paper's argument, but the paper may add assumptions or extra steps.
extends
The paper goes beyond the formal theorem; the theorem is a base layer rather than the whole result.
uses
The paper appears to rely on the theorem as machinery.
contradicts
The paper's claim conflicts with a theorem or certificate in the canon.
unclear
Pith found a possible connection, but the passage is too broad, indirect, or ambiguous to say the theorem truly supports the claim.

Reference graph

Works this paper leans on

78 extracted references · 78 canonical work pages

  1. [1]

    Eris®E5 Studio Monitor

    2025. Eris®E5 Studio Monitor. https://intl.presonus.com/products/eris-e5-studio-monitor?srsltid=AfmBOopFGpJKhO6xHwJ_ iRWlSQZsIOf9Kiw8Wv5WjZ0mKTitWmFcpOnO

    work page 2025

  2. [2]

    GRAS 45BC KEMAR Head and Torso with Mouth Simulato

    2025. GRAS 45BC KEMAR Head and Torso with Mouth Simulato. https://www.grasacoustics.com/products/head-torso-simulators-kemar/kemar- non-configured/product/749-45bc

    work page 2025

  3. [3]

    Infineon IM73D122V01: Ultra-low noise high sensitivity digital XENSIVTM MEMS microphone

    2025. Infineon IM73D122V01: Ultra-low noise high sensitivity digital XENSIVTM MEMS microphone. https://www.infineon.com/assets/row/public/ documents/24/49/infineon-im73d122-datasheet-en.pdf?fileId=8ac78c8c83cd3081018409cafdfb46db

    work page 2025

  4. [4]

    Microsoft Hololens

    2025. Microsoft Hololens. https://learn.microsoft.com/en-us/hololens/

    work page 2025

  5. [5]

    Orange Pi Zero2 Module

    2025. Orange Pi Zero2 Module. http://www.orangepi.org

    work page 2025

  6. [6]

    Ray-ban Meta

    2025. Ray-ban Meta. https://www.ray-ban.com/usa/ray-ban-meta-ai-glasses

    work page 2025

  7. [7]

    Rokid Glasses

    2025. Rokid Glasses. https://glasses.rokid.com

    work page 2025

  8. [8]

    Sonion Voice Picking Unit VPU14aa01

    2025. Sonion Voice Picking Unit VPU14aa01. https://www.ariat-tech.com/parts/VPU14AA01?srsltid= AfmBOoqPZF0ySUfKmsHCBWudGqWET3zyMNi78FVCvMtjfjzo8pOYW5_n

    work page 2025

  9. [9]

    Xreal One Pro

    2025. Xreal One Pro. https://www.xreal.com/one-pro

    work page 2025

  10. [10]

    Muhammad Ejaz Ahmed, Il-Youp Kwak, Jun Ho Huh, Iljoo Kim, Taekkyung Oh, and Hyoungshick Kim. 2020. Void: A fast and light voice liveness detection system. In29th USENIX Security Symposium (USENIX Security 20). 2685–2702

    work page 2020

  11. [11]

    Zahid Akhtar, Christian Michelon, and Gian Luca Foresti. 2014. Liveness detection for biometric authentication in mobile applications. In2014 International Carnahan Conference on Security Technology (ICCST). IEEE, 1–6

    work page 2014

  12. [12]

    Pierre Badin and Antoine Serrurier. 2006. Three-dimensional modeling of speech organs: Articulatory data and models. InTechnical Committee of Psychological and Physiological Acoustics, Vol. 36. The Acoustical Society of Japan, 421–426

    work page 2006

  13. [13]

    Logan Blue, Hadi Abdullah, Luis Vargas, and Patrick Traynor. 2018. 2ma: Verifying voice commands via two microphone authentication. In Proceedings of the 2018 on Asia Conference on Computer and Communications Security. 89–100

    work page 2018

  14. [14]

    Dutliff Boshoff and Gerhard P Hancke. 2025. A classifications framework for continuous biometric authentication (2018–2024).Computers & Security 150 (2025), 104285

    work page 2025

  15. [15]

    Bruno Bouchard, Kevin Bouchard, and Abdenour Bouzouane. 2020. A smart cooking device for assisting cognitively impaired users.Journal of reliable intelligent environments6, 2 (2020), 107–125. Manuscript submitted to ACM AuthGlass: Enhancing Voice Authentication on Smart Glasses via Air-Bone Acoustic Features 13

    work page 2020

  16. [16]

    Pan Chan, Tzipora Halevi, and Nasir Memon. 2015. Glass otp: Secure and convenient user authentication on google glass. InInternational Conference on Financial Cryptography and Data Security. Springer, 298–308

    work page 2015

  17. [17]

    Wan-Jung Chang, Liang-Bi Chen, and Yu-Zung Chiou. 2018. Design and implementation of a drowsiness-fatigue-detection system based on wearable smart glasses to increase road safety.IEEE Transactions on Consumer Electronics64, 4 (2018), 461–469

    work page 2018

  18. [18]

    2018.A two-layer authentication using voiceprint for voice assistants

    Yun-Tai Chang. 2018.A two-layer authentication using voiceprint for voice assistants. Ph. D. Dissertation

    work page 2018

  19. [19]

    Si Chen, Kui Ren, Sixu Piao, Cong Wang, Qian Wang, Jian Weng, Lu Su, and Aziz Mohaisen. 2017. You can hear but you cannot steal: Defending against voice impersonation attacks on smartphones. In2017 IEEE 37th international conference on distributed computing systems (ICDCS). IEEE, 183–195

    work page 2017

  20. [20]

    Serife Kucur Ergünay, Elie Khoury, Alexandros Lazaridis, and Sébastien Marcel. 2015. On the vulnerability of speaker verification to realistic voice spoofing. In2015 IEEE 7th international conference on biometrics theory, applications and systems (BTAS). IEEE, 1–6

    work page 2015

  21. [21]

    Huan Feng, Kassem Fawaz, and Kang G Shin. 2017. Continuous authentication for voice assistants. InProceedings of the 23rd Annual International Conference on Mobile Computing and Networking. 343–355

    work page 2017

  22. [22]

    Rainhard Dieter Findling, Tahmid Quddus, and Stephan Sigg. 2019. Hide my gaze with EOG! towards closed-eye gaze gesture passwords that resist observation-attacks with electrooculography in smart glasses. InProceedings of the 17th international conference on advances in mobile computing & multimedia. 107–116

    work page 2019

  23. [23]

    Eira Friström, Elias Lius, Niki Ulmanen, Paavo Hietala, Pauliina Kärkkäinen, Tommi Mäkinen, Stephan Sigg, and Rainhard Dieter Findling. 2019. Free-form gaze passwords from cameras embedded in smart glasses. InProceedings of the 17th International Conference on Advances in Mobile Computing & Multimedia. 136–144

    work page 2019

  24. [24]

    Bhanuka Gamage. 2024. AI-Enabled Smart Glasses for People with Severe Vision Impairments.ACM SIGACCESS Accessibility and Computing137 (2024), 1–1

    work page 2024

  25. [25]

    Yaroslav Ganin and Victor Lempitsky. 2015. Unsupervised domain adaptation by backpropagation. InInternational conference on machine learning. PMLR, 1180–1189

    work page 2015

  26. [26]

    Yang Gao, Yincheng Jin, Jagmohan Chauhan, Seokmin Choi, Jiyang Li, and Zhanpeng Jin. 2021. Voice in ear: Spoofing-resistant and passphrase- independent body sound authentication.Proceedings of the ACM on Interactive, Mobile, Wearable and Ubiquitous Technologies5, 1 (2021), 1–25

    work page 2021

  27. [27]

    Kaiyi Guo, Tianyu Wu, Yang Gao, Qian Zhang, and Dong Wang. 2025. EchoTouch: Low-power Face-touching Behavior Recognition Using Active Acoustic Sensing on Glasses.Proc. ACM Interact. Mob. Wearable Ubiquitous Technol.9, 2, Article 31 (June 2025), 33 pages. doi:10.1145/3729481

    work page doi:10.1145/3729481 2025

  28. [28]

    Kaiyi Guo, Qian Zhang, and Dong Wang. 2025. EchoBreath: Continuous Respiratory Behavior Recognition in the Wild via Acoustic Sensing on Smart Glasses. InProceedings of the 2025 CHI Conference on Human Factors in Computing Systems (CHI ’25). Association for Computing Machinery, New York, NY, USA, Article 348, 21 pages. doi:10.1145/3706598.3714171

    work page doi:10.1145/3706598.3714171 2025

  29. [29]

    Jibo He, Jake Ellis, William Choi, and Pingfeng Wang. 2015. Driving While Reading Using Google Glass Versus Using a Smartphone: Which is More Distracting to Driving Performance?. InDriving Assessment Conference, Vol. 8. University of Iowa

    work page 2015

  30. [30]

    Chenpei Huang, Hui Zhong, Pavana Prakash, Dian Shi, Xu Yuan, and Miao Pan. 2025. Eve said yes: Airbone authentication for head-wearable smart voice assistant.IEEE Transactions on Mobile Computing(2025)

    work page 2025

  31. [31]

    Masaya Inoue and Kazuya Murao. 2024. User Authentication Method for Smart Glasses using Gaze Information of Registered Known Images and AI-generated Unknown Images. InCompanion of the 2024 on ACM International Joint Conference on Pervasive and Ubiquitous Computing. 525–530

    work page 2024

  32. [32]

    MD Rasel Islam, Doyoung Lee, Liza Suraiya Jahan, and Ian Oakley. 2018. Glasspass: Tapping gestures to unlock smart glasses. InProceedings of the 9th Augmented Human International Conference. 1–8

    work page 2018

  33. [33]

    Kaito Isobe and Kazuya Murao. 2021. Person-identification methodusing active acoustic sensing applied to nose. InProceedings of the 2021 ACM International Symposium on Wearable Computers. 138–140

    work page 2021

  34. [34]

    Koji Iwano, Taro Miyazaki, and Sadaoki Furui. 2005. Multimodal speaker verification using ear image features extracted by PCA and ICA. In International Conference on Audio-and Video-Based Biometric Person Authentication. Springer, 588–596

    work page 2005

  35. [35]

    Madhu R Kamble, Hardik B Sailor, Hemant A Patil, and Haizhou Li. 2020. Advances in anti-spoofing: from the perspective of ASVspoof challenges. APSIPA Transactions on Signal and Information Processing9 (2020), e2

    work page 2020

  36. [36]

    Lawrence George Kersta. 1962. Voiceprint identification.The Journal of the Acoustical Society of America34, 5_Supplement (1962), 725–725

    work page 1962

  37. [37]

    Smita Khade, Swati Ahirrao, Shraddha Phansalkar, Ketan Kotecha, Shilpa Gite, and Sudeep D Thepade. 2021. Iris liveness detection for biometric authentication: A systematic literature review and future directions.Inventions6, 4 (2021), 65

    work page 2021

  38. [38]

    Azizuddin Khan and Gyan Prakash. 2017. Design and implementation of smart glass with voice detection capability to help visually impaired people. International Journal of MC Square Scientific Research9, 3 (2017), 54–59

    work page 2017

  39. [39]

    smart glasses

    Sunwook Kim, Maury A Nussbaum, and Joseph L Gabbard. 2016. Augmented reality “smart glasses” in the workplace: industry perspectives and challenges for worker safety and health.IIE transactions on occupational ergonomics and human factors4, 4 (2016), 253–258

    work page 2016

  40. [40]

    Huining Li, Chenhan Xu, Aditya Singh Rathore, Zhengxiong Li, Hanbin Zhang, Chen Song, Kun Wang, Lu Su, Feng Lin, Kui Ren, et al . 2020. VocalPrint: Exploring a resilient and secure voice authentication via mmWave biometric interrogation. InProceedings of the 18th Conference on Embedded Networked Sensor Systems. 312–325

    work page 2020

  41. [41]

    Jingjin Li, Chao Chen, Mostafa Rahimi Azghadi, Hossein Ghodosi, Lei Pan, and Jun Zhang. 2023. Security and privacy problems in voice assistant applications: A survey.Computers & Security134 (2023), 103448. Manuscript submitted to ACM 14 Xu et al

    work page 2023

  42. [42]

    Ke Li, Devansh Agarwal, Ruidong Zhang, Vipin Gunda, Tianjun Mo, Saif Mahmud, Boao Chen, François Guimbretiěre, and Cheng Zhang. 2024. SonicID: User Identification on Smart Glasses with Acoustic Sensing.Proceedings of the ACM on Interactive, Mobile, Wearable and Ubiquitous Technologies8, 4 (2024), 1–27

    work page 2024

  43. [43]

    Yue Li, Xueru Gao, Qipeng Song, Yao Wang, Pin Lyu, and Haibin Zhang. 2024. BoneAuth: A Bone Conduction-Based Voice Liveness Authentication for Voice Assistants.IEEE Internet of Things Journal(2024)

    work page 2024

  44. [44]

    Yung-Hui Li and Po-Jen Huang. 2017. An accurate and efficient user authentication mechanism on smart glasses based on iris recognition.Mobile Information Systems2017, 1 (2017), 1281020

    work page 2017

  45. [45]

    Zhuohang Li, Cong Shi, Tianfang Zhang, Yi Xie, Jian Liu, Bo Yuan, and Yingying Chen. 2021. Robust Detection of Machine-induced Audio Attacks in Intelligent Audio Systems with Microphone Array. InProceedings of the 2021 ACM SIGSAC Conference on Computer and Communications Security (Virtual Event, Republic of Korea)(CCS ’21). Association for Computing Machi...

    work page doi:10.1145/3460120.3484755 2021

  46. [46]

    Hyunchul Lim, Guilin Hu, Richard Jin, Hao Chen, Ryan Mao, Ruidong Zhang, and Cheng Zhang. 2023. C-Auth: Exploring the Feasibility of Using Egocentric View of Face Contour for User Authentication on Glasses. InProceedings of the 2023 ACM International Symposium on Wearable Computers. 6–10

    work page 2023

  47. [47]

    Rui Liu, Cory Cornelius, Reza Rawassizadeh, Ronald Peterson, and David Kotz. 2018. Vocal Resonance: Using Internal Body Voice for Wearable Authentication.Proc. ACM Interact. Mob. Wearable Ubiquitous Technol.2, 1, Article 19 (March 2018), 23 pages. doi:10.1145/3191751

    work page doi:10.1145/3191751 2018

  48. [48]

    Isaias Majil, Mau-Tsuen Yang, and Sophia Yang. 2022. Augmented reality based interactive cooking guide.Sensors22, 21 (2022), 8290

    work page 2022

  49. [49]

    Yan Meng, Jiachun Li, Matthew Pillari, Arjun Deopujari, Liam Brennan, Hafsah Shamsie, Haojin Zhu, and Yuan Tian. 2022. Your microphone array retains your identity: A robust voice liveness detection system for smart speakers. In31st USENIX Security Symposium (USENIX Security 22). 1077–1094

    work page 2022

  50. [50]

    Johannes Meyer, Adrian Frank, Thomas Schlebusch, and Enkelejda Kasneci. 2022. U-har: A convolutional approach to human activity recognition combining head and eye movements for context-aware smart glasses.Proceedings of the ACM on Human-Computer Interaction6, ETRA (2022), 1–19

    work page 2022

  51. [51]

    AN Nithyaa, R Premkumar, N Sudhandirapriya, and R Durga. 2023. Eye Glasses for the Visually Challenged Using Artificial Intelligence Application. InInternational Conference on Soft Computing and Signal Processing. Springer, 347–360

    work page 2023

  52. [52]

    Francesca Palermo, Luca Casciano, Lokmane Demagh, Aurelio Teliti, Niccolò Antonello, Giacomo Gervasoni, Hazem Hesham Yousef Shalby, Marco Brando Paracchini, Simone Mentasti, Hao Quan, et al. 2025. Advancements in Context Recognition for Edge Devices and Smart Eyewear: Sensors and Applications.IEEE Access(2025)

    work page 2025

  53. [53]

    Ge Peng, Gang Zhou, David T Nguyen, Xin Qi, Qing Yang, and Shuangquan Wang. 2016. Continuous authentication with touch behavioral biometrics and voice on wearable glasses.IEEE transactions on human-machine systems47, 3 (2016), 404–416

    work page 2016

  54. [54]

    Yue Qin, Chun Yu, Zhaoheng Li, Mingyuan Zhong, Yukang Yan, and Yuanchun Shi. 2021. Proximic: Convenient voice activation via close-to-mic speech detected by a single microphone. InProceedings of the 2021 CHI Conference on Human Factors in Computing Systems. 1–12

    work page 2021

  55. [55]

    Selvi Rajendran, Padmaveni Krishnan, and D

    P. Selvi Rajendran, Padmaveni Krishnan, and D. John Aravindhar. 2020. Design and Implementation of Voice Assisted Smart Glasses for Visually Impaired People Using Google Vision API. In2020 4th International Conference on Electronics, Communication and Aerospace Technology (ICECA). 1221–1224. doi:10.1109/ICECA49313.2020.9297553

    work page doi:10.1109/iceca49313.2020.9297553 2020

  56. [56]

    P Selvi Rajendran, Padmaveni Krishnan, and D John Aravindhar. 2020. Design and implementation of voice assisted smart glasses for visually impaired people using google vision api. In2020 4th International Conference on Electronics, Communication and Aerospace Technology (ICECA). IEEE, 1221–1224

    work page 2020

  57. [57]

    Md Sahidullah, Dennis Alexander Lehmann Thomsen, Rosa Gonzalez Hautamäki, Tomi Kinnunen, Zheng-Hua Tan, Robert Parts, and Martti Pitkänen. 2017. Robust voice liveness detection and speaker verification using throat microphones.IEEE/ACM Transactions on Audio, Speech, and Language Processing26, 1 (2017), 44–56

    work page 2017

  58. [58]

    Stefan Schneegass, Youssef Oualil, and Andreas Bulling. 2016. SkullConduct: Biometric user identification on eyewear computers using bone conduction through the skull. InProceedings of the 2016 CHI Conference on Human Factors in Computing Systems. 1379–1384

    work page 2016

  59. [59]

    Hyejin Shin, Jun Ho Huh, Bum Jun Kwon, Iljoo Kim, Eunyong Cheon, HongMin Kim, Choong-Hoon Lee, and Ian Oakley. 2024. SkullID: Through-skull sound conduction based authentication for smartglasses. InProceedings of the 2024 CHI Conference on Human Factors in Computing Systems. 1–19

    work page 2024

  60. [60]

    Sayaka Shiota, Fernando Villavicencio, Junichi Yamagishi, Nobutaka Ono, Isao Echizen, and Tomoko Matsui. 2015. Voice liveness detection algorithms based on pop noise caused by human breath for automatic speaker verification. InINTERSPEECH 2015 16th Annual Conference of the International Speech Communication Association. International Speech Communication ...

    work page 2015

  61. [61]

    Ashique T P, Hridya Girish, Ivine P Sabu, Thomas Raju, and Anoop V. 2025. AI-Powered Smart Glasses for Real-Time Speech-to-Text Transcription. In2025 4th International Conference on Advances in Computing, Communication, Embedded and Secure Systems (ACCESS). 442–445. doi:10.1109/ ACCESS65134.2025.11135871

    work page arXiv 2025

  62. [62]

    Suresh Veesa and Madhusudan Singh. 2025. Deep learning countermeasures for detecting replay speech attacks: a review.International Journal of Speech Technology28, 1 (2025), 39–51

    work page 2025

  63. [63]

    Qian Wang, Xiu Lin, Man Zhou, Yanjiao Chen, Cong Wang, Qi Li, and Xiangyang Luo. 2019. VoicePop: A Pop Noise based Anti-spoofing System for Voice Authentication on Smartphones. InIEEE INFOCOM 2019 - IEEE Conference on Computer Communications. 2062–2070. doi:10.1109/INFOCOM. 2019.8737422

    work page doi:10.1109/infocom 2019

  64. [64]

    Yao Wang, Wandong Cai, Tao Gu, Wei Shao, Yannan Li, and Yong Yu. 2019. Secure your voice: An oral airflow-based continuous liveness detection for voice assistants.Proceedings of the ACM on interactive, mobile, wearable and ubiquitous technologies3, 4 (2019), 1–28. Manuscript submitted to ACM AuthGlass: Enhancing Voice Authentication on Smart Glasses via A...

    work page 2019

  65. [65]

    Zeyu Wang, Yuanchun Shi, Yuntao Wang, Yuchen Yao, Kun Yan, Yuhan Wang, Lei Ji, Xuhai Xu, and Chun Yu. 2024. G-VOILA: gaze-facilitated information querying in daily scenarios.Proceedings of the ACM on Interactive, Mobile, Wearable and Ubiquitous Technologies8, 2 (2024), 1–33

    work page 2024

  66. [66]

    Hello, It’s Me

    Emily Wenger, Max Bronckers, Christian Cianfarani, Jenna Cryan, Angela Sha, Haitao Zheng, and Ben Y Zhao. 2021. " Hello, It’s Me": Deep Learning-based Speech Synthesis Attacks in the Real World. InProceedings of the 2021 ACM SIGSAC Conference on Computer and Communications Security. 235–251

    work page 2021

  67. [67]

    Christian Winkler, Jan Gugenheimer, Alexander De Luca, Gabriel Haas, Philipp Speidel, David Dobbelstein, and Enrico Rukzio. 2015. Glass unlock: Enhancing security of smartphone unlocking through leveraging a private near-eye display. InProceedings of the 33rd annual acm conference on human factors in computing systems. 1407–1410

    work page 2015

  68. [68]

    Chen Yan, Xiaoyu Ji, Kai Wang, Qinhong Jiang, Zizhi Jin, and Wenyuan Xu. 2022. A survey on voice assistant security: Attacks and countermeasures. Comput. Surveys55, 4 (2022), 1–36

    work page 2022

  69. [69]

    Chen Yan, Yan Long, Xiaoyu Ji, and Wenyuan Xu. 2019. The catcher in the field: A fieldprint based spoofing detection for text-independent speaker verification. InProceedings of the 2019 ACM SIGSAC Conference on Computer and Communications Security. 1215–1229

    work page 2019

  70. [70]

    Jiangyan Yi, Chenglong Wang, Jianhua Tao, Xiaohui Zhang, Chu Yuan Zhang, and Yan Zhao. 2023. Audio deepfake detection: A survey.arXiv preprint arXiv:2308.14970(2023)

    work page arXiv 2023

  71. [71]

    Linghan Zhang, Sheng Tan, Yingying Chen, and Jie Yang. 2022. A Continuous Articulatory-Gesture-Based Liveness Detection for Voice Authentication on Smart Devices.IEEE Internet of Things Journal9, 23 (2022), 23320–23331. doi:10.1109/JIOT.2022.3199995

    work page doi:10.1109/jiot.2022.3199995 2022

  72. [72]

    Linghan Zhang, Sheng Tan, and Jie Yang. 2017. Hearing your voice is not enough: An articulatory gesture based liveness detection for voice authentication. InProceedings of the 2017 ACM SIGSAC Conference on Computer and Communications Security. 57–71

    work page 2017

  73. [73]

    Linghan Zhang, Sheng Tan, Jie Yang, and Yingying Chen. 2016. Voicelive: A phoneme localization based liveness detection for voice authentication on smartphones. InProceedings of the 2016 ACM SIGSAC Conference on Computer and Communications Security. 1080–1091

    work page 2016

  74. [74]

    Rui Zhang, Zheng Yan, Xuerui Wang, and Robert H Deng. 2022. Livoauth: Liveness detection in voiceprint authentication with random challenges and detection modes.IEEE Transactions on Industrial Informatics19, 6 (2022), 7676–7688

    work page 2022

  75. [75]

    Tianfang Zhang, Qiufan Ji, Zhengkun Ye, Md Mojibur Rahman Redoy Akanda, Ahmed Tanvir Mahdad, Cong Shi, Yan Wang, Nitesh Saxena, and Yingying Chen. 2024. SAFARI: Speech-Associated Facial Authentication for AR/VR Settings via Robust VIbration Signatures. InProceedings of the 2024 on ACM SIGSAC Conference on Computer and Communications Security. 153–167

    work page 2024

  76. [76]

    Yongtuo Zhang, Wen Hu, Weitao Xu, Chun Tung Chou, and Jiankun Hu. 2018. Continuous authentication using eye movement response of implicit visual stimuli.proceedings of the acm on interactive, mobile, wearable and ubiquitous technologies1, 4 (2018), 1–22

    work page 2018

  77. [77]

    Zhan Zhang, Enze Bai, Aram Stepanian, Swathi Jagannath, and Sun Young Park. 2025. Touchless interaction for smart glasses in emergency medical services: user needs and experiences.International Journal of Human–Computer Interaction41, 5 (2025), 2984–3003

    work page 2025

  78. [78]

    Chu-Xiao Zuo, Zhi-Jun Jia, and Wu-Jun Li. 2024. AdvTTS: Adversarial Text-to-Speech Synthesis Attack on Speaker Identification Systems. In ICASSP 2024-2024 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP). IEEE, 4840–4844. Received 20 February 2007; revised 12 March 2009; accepted 5 June 2009 Manuscript submitted to ACM

    work page 2024