This paper presents Signly.uz, a mobile learning platform for Uzbek and American Sign Language (UzSL/ASL) built around a gamified mobile experience. The app runs on Flutter with a Firebase backend, and TensorFlow Lite handles offline recognition for a subset of the 500+ signs in our companion dataset [6]. Instead of treating sign-language technology purely as a real-time translation aid, we built in gesture-based games, interactive flashcards, multimedia learning material, quizzes, and an XP/badge progress system – an approach supported by prior work showing gamification boosts engagement [2]. We piloted the app in a classroom setting with 10 university coursemates, who tried out the vocabulary activities and recognition features and gave feedback afterward. Reactions were generally positive: participants reported increased motivation to practice, and the XP leaderboard and quiz mode came up most often as favorite features. We also saw gesture misclassification on lower-end Android devices and occasional lag during video rendering. We present Signly.uz as an educational extension of sign-language recognition technology rather than a validated learning intervention, the sample was small and this was not a controlled effectiveness trial.
Uzbek Sign Language; American Sign Language; mobile learning; gamification; TensorFlow Lite; Flutter; sign language recognition; inclusive education; offline learning
[1] Ergashev, J., Eshimov, M., & Rakhmatova, M. (2025). Sign Language Translation
Model. B.S. thesis, Central Asian University, Tashkent, Uzbekistan.
[2] Hamari, J., Koivisto, J., & Sarsa, H. (2014). Does gamification work? A literature
review of empirical studies. In Proceedings of the 47th Hawaii International Conference
on System Sciences (HICSS) (pp. 3025–3034). IEEE.
[3] Woll, B., & Morgan, G. (2012). Language learning and language teaching in deaf
education. Cambridge University Press.
[4] Lugaresi, C., Tang, J., Nash, H., et al. (2019). MediaPipe: A framework for building
perception pipelines. arXiv:1906.08172.
[5] Cao, Z., Hidalgo, G., Simon, T., Wei, S.-E., & Sheikh, Y. (2021). OpenPose: Realtime
multi-person 2D pose estimation using part affinity fields. IEEE Transactions on Pattern
Analysis and Machine Intelligence, 43(1), 172–186.