2025
doi: 10.12720/jait.16.10.1379-1387
An Automated Visual Acuity Test System Using Vietnamese Speech Recognition for Answer Selection
Pham Hoang Minh*
,
Nguyen Duc Thanh ,
Pham Hong Duong
- Institute of Materials Science, Vietnam Academy of Science and Technology, Hanoi, Vietnam
- Manuscript receivedJune 11, 2025
- revisedJuly 1, 2025
- acceptedJuly 17, 2025
- publishedOctober 14, 2025
Abstract
Visual Acuity (VA) testing traditionally requires an ophthalmologist, limiting accessibility in non-clinical settings. This paper presents an automated VA test system designed for precise vision assessment without professional supervision, which leverages the Early Treatment Diabetic Retinopathy Study (ETDRS) eye chart for higher accuracy compared to the conventional Snellen chart. The system uses a computer with an Liquid Crystal Display (LCD) monitor and incorporates automated scoring with 0.02 LogMAR precision. To facilitate remote operation, we implement speech recognition in Vietnamese via a microphone, utilizing Azure Speech API, which is enhanced with a correction function and noise classification for improved accuracy. An experiment with 80 participants (N = 80) demonstrated a speech recognition accuracy of 93.1%, with a mean response time of 4.6 s per optotype. The VA scores from our system closely matched those from standard printed ETDRS charts, with 95.6% of measurements differing by ≤0.1 LogMAR. Our automated VA test system provides a reliable, low-cost solution for vision assessment in non-clinical environments, combining high accuracy with user-friendly remote operation.
Keywords
- automated visual acuity test
- automated Early Treatment Diabetic Retinopathy Study (ETDRS) test
- speech recognition
- isolated letter recognition
How to Cite
Pham Hoang Minh, Nguyen Duc Thanh, and Pham Hong Duong, "An Automated Visual Acuity Test System Using Vietnamese Speech Recognition for Answer Selection," Journal of Advances in Information Technology, vol. 16, no. 10, pp. 1379-1387, 2025. https://doi.org/10.12720/jait.16.10.1379-1387
Copyright & License
Copyright © 2025 by the authors. This is an open access article distributed under the Creative Commons Attribution License which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited (CC BY 4.0).
