Real-Time Kazakh Sign Language Alphabet Recognition Using SVM and YOLOv8n

Ramazan Duisenbek, Tamara Kokenovna Zhukabayeva

2025

Closed access Metadata (CSL-JSON)
DOI 10.1109/sist61657.2025.11139187
Data confidence 70 medium
How this score is calculated

It is built from checkable properties of the record itself: whether the article has a DOI, how many independent sources confirmed its data, whether the required fields are filled in, and how recent the latest confirmation is. Disagreement between sources and a missing identifier lower it. A value of 80 or above means the record was published without further checking; 60 to 79 means it is published at reduced confidence, with a manual check advisable. A record verified by an editor by hand is never pushed below 90 by a recalculation. This scores our data about the article, not the article itself.

Abstract

The development of Kazakh Sign Language (KSL) recognition is an important step in enhancing opportunities and communication for the deaf and hard-hearing communities. This work focuses on the recognition of the Kazakh Sign Language alphabet using YOLOv8n for real-time gesture recognition and a support vector machine (SVM) for classifying gesture images. SVM achieved an accuracy rate of 99% for classification and YOLOv8n detected all gestures in real-time. This paper demonstrates the first use of YOLOv8n model to KSL alphabet recognition, making a significant contribution to this field. This research gives a foundation for upcoming works and shows a need in expanding datasets with dynamic gestures and recognizing words with full sentences The results of this work not only advance sign language recognition systems in Kazakhstan but also demonstrate the potentials of broader applications in other sign languages. This research represents an important step toward closing gaps for the deaf community.