Home / Current Issue / Paper 1704651
SignAssist: Sign Language Interpreter using Deep Learning
Subject area: Science,Engineering and Technology · Area of research: Deep Learning, Human-Machine Interaction
Abstract
Effective communication is essential for individuals to express their ideas and emotions. However, persons with speech or hearing disabilities often face significant communication barriers. To address this issue, deep learning models, specifically LSTM and GRU, are proposed to recognize and translate signs from isolated American Sign Language (ASL) video frames. In this research, transfer learning and data augmentation techniques are utilized to develop a deep learning model for the ASL dataset. The proposed models achieve up to 95% accuracy in recognizing signs from ASL datasets. This research aims to develop a more natural and efficient way of communication for persons with hearing impairments and promote collaboration with people not trained in sign language. Overall, this study demonstrates the potential of deep learning models to reduce communication barriers and promote inclusivity.
Keywords
American Sign Language, SignAssist, Deep Learning, Sign Recognition, Gesture Recognition
How to cite this paper
@article{1704651,
author = {Karan Kharbanda, Utsav Sachdeva, Prof. Dr. Anu Rathee},
title = {SignAssist: Sign Language Interpreter using Deep Learning},
journal = {Iconic Research And Engineering Journals},
year = {2023},
volume = {6},
number = {12},
pages = {387-393},
issn = {2456-8880},
url = {https://www.irejournals.com/formatedpaper/1704651.pdf},
abstract = {Effective communication is essential for individuals to express their ideas and emotions. However, persons with speech or hearing disabilities often face significant communication barriers. To address this issue, deep learning models, specifically LSTM and GRU, are proposed to recognize and translate signs from isolated American Sign Language (ASL) video frames. In this research, transfer learning and data augmentation techniques are utilized to develop a deep learning model for the ASL dataset. The proposed models achieve up to 95% accuracy in recognizing signs from ASL datasets. This research aims to develop a more natural and efficient way of communication for persons with hearing impairments and promote collaboration with people not trained in sign language. Overall, this study demonstrates the potential of deep learning models to reduce communication barriers and promote inclusivity.},
keywords = {American Sign Language, SignAssist, Deep Learning, Sign Recognition, Gesture Recognition},
month = {June},
}