Convolutional Neural Network-Based Real-Time Kannada Sign Language Detection

The development of a real-time sign language translator represents a crucial advancement toward enhancing communication between the deaf community and the general public. Almost three hundred sign languages are used worldwide, each of which reflects the customs and cultures of its speakers. The distinctiveness of the speech-impaired community in Karnataka is greatly influenced by Kannada sign language, which, like other regional sign languages, is influenced by linguistic accommodation. Real-time detection has not been thoroughly researched, despite the fact that few studies have attempted to predict Kannada sign language using static photos. By providing better accuracy in hand gesture recognition and interpretation in real-time within the context of Kannada sign language, the suggested methodology seeks to fill this gap. In this study, a method for understanding Kannada sign language is presented. The proposed work offers a brand-new technique for utilizing CNN to classify and identify Kannada language signs. The objective of the proposed system is to enhance real-time recognition software, enabling its versatile application in any location.