Copyright © 2026 Authors retain the copyright of this article. This article 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.

@article{192526,
        author = {Sheetal Patil and Dr Amol Kadam and Ashwini Patil},
        title = {Designing an AI-powered solution to recognize and classify dynamic hand gestures in indian sign language, aimed at enhancing communication for the hearing and speech impaired},
        journal = {International Journal of Innovative Research in Technology},
        year = {2026},
        volume = {12},
        number = {9},
        pages = {1369-1375},
        issn = {2349-6002},
        url = {https://ijirt.org/article?manuscript=192526},
        abstract = {In recent years, communication accessibility for individuals who are mute and deaf has gained significant attention, particularly through the use of sign language. Indian Sign Language (ISL), being the native sign language for millions, remains underrepresented in mainstream technological solutions. This project proposes an intelligent, real-time system for dynamic hand gesture detection and classification, specifically tailored for ISL, utilizing advanced deep learning techniques. The objective of this study is to develop a robust system that can accurately recognize and translate dynamic hand gestures those involving motion over time into readable or spoken text, thereby bridging gaps between hearing-impaired individuals. The system captures live video input through a webcam or camera sensor, detects hand movements in real time, and interprets gestures using a Convolutional Neural Network (CNN) combined with a Recurrent Neural Network (RNN), particularly Long Short-Term Memory (LSTM) layers, to model temporal dependencies in gesture sequences. The dataset used includes labeled dynamic ISL gesture sequences collected from diverse users to ensure generalizability. Preprocessing steps such as background subtraction, skin-color segmentation, and hand region normalization are applied to improve detection accuracy. The deep learning model is trained to classify gestures into predefined classes corresponding to words or phrases in ISL. Experimental results demonstrate high accuracy, speed, and reliability of the system in recognizing dynamic gestures across various environments and lighting conditions. Furthermore, the system is designed to be easily deployable on mobile or embedded platforms, enhancing its usability in real-world applications such as education, public services, and assistive tools.},
        keywords = {Assistive Technology, Convolutional Neural Network (CNN), Deaf and Mute Communication, Deep Learning, Human computer interaction},
        month = {February},
        }