Bangla Sign Language Recognition and Sentence Building Using Deep Learning

Abstract

Modern age being the era of Information technology, it would not have come this far without the piled up data or information. Whereas communication is the basis of collecting or gathering data or information, almost 5% of the world's population is not blessed with the ability of verbal communication [1]. For deaf and dumb people lacking the ability of verbal communication, sign language is the solution. Sign language varies from the verbal language in every form and rule. This creates a gap between people conversing in verbal language and those communicating in sign language. Verbal languages are easy to interpret for having a common rule-following but sign language differs from region to region. This hampers the communication between normal people and those interacting in sign languages. Human to human interpretation is tough because of the enriched word wise signs and vocabs. To eradicate this issue, we are proposing a machine-based approach for training and detecting the Bangla Sign Language. Our aim is to train the system with enough samples containing different signs used in Bangla Sign Language. In this research, we are using the Convolutional Neural Network (CNN) for training each individual sign. In addition to working as a medium of communication between the deaf and mute with the remaining society, this approach would also serve as a tool for the hearing deprived to learn and use the sign language properly. Moreover, this would also come to assistance for anyone willing to learn or develop sign language or wishes to work with those with special needs of using sign language.

Real-Time Arabic Sign Language Recognition Using a Hybrid Deep Learning Model

“…Despite challenges and limitations related to the dataset’s size and the full representation of sign language diversity, the study achieved a training accuracy of 87%, indicating the model’s ability to accurately classify the signs. In [ 29 ], the authors proposed an approach to train and detect Bengali Sign Language using deep learning. The authors used a CNN classifier to train each sign individually.…”

Using Deep Learning in Sign Language Translation to Text

“…It is used by (Wadhawan and Kumar, 2020), (Abdul et al, 2021), (Zhang et al, 2019), (Park et al, 2021), (Ferreira et al, 2019), , (Cihan Camgoz et al, 2017), (Sridhar et al, 2020), (Alam et al, 2021(Gangrade andBharti, 2020), (Breland et al, 2021), , (Orbay and Akarun, 2020) and (Shurid et al, 2020). CNN is is a tool for extracting information from images (Park et al, 2021), used in image processing and recognition (Alam et al, 2021) and composed of convolutional layers followed by additional layers (Wadhawan and Kumar, 2020).…”

“…The various studies under sign language recognition containing its content, description and respective references are shown in Table 2. (Abdul et al, 2021), (Sridhar et al, 2020), (Alam et al, 2021), (Breland et al, 2021), (Shurid et al, 2020) Videos Videos as data sets to recognized sign language (Abdul et al, 2021), (Wadhawan and Kumar, 2020), (Yu et al, 2019), (Cihan Camgoz et al, 2017), (Sridhar et al, 2020), (Alam et al, 2021) Wearable sensors…”

“…The study is about sign language translation into text; this technology is for those who are experiencing difficulty in communicating i.e., those who have speech disorders, hearing impairment, and are deaf (Fang et al, 2017) (Ferreira et al, 2019) (Sridhar et al, 2020) (Alam et al, 2021) (Orbay and Akarun, 2020) (Shurid et al, 2020) (Hou et al, 2019)]. Sign language translation into text will help them communicate with other people as they can express their feelings thoroughly without the concern of misunderstanding them, or not understanding them at all.…”

“…This section cites a variety of journal(Hu et al, 2018) (Yu et al, 2019) (Gangrade and Bharti, 2020) (Breland et al, 2021) and conference(Fang et al, 2017) (Zhang et al, 2019) (Cihan Camgoz et al, 2017) (Sridhar et al, 2020) (Orbay and Akarun, 2020) (Shurid et al, 2020) (Hou et al, 2019) articles to show how they appear in the references section. It also cites different research papers(Wadhawan and Kumar, 2020)(Alam et al, 2021) and proposal studies(Abdul et al, 2021) (Park et al, 2021) (Jin et al, 2021 (Ferreira et al, 2019) …”