Sign language detection and conversion to readable Bengali words using BdSL

Abstract

In this era of modernization, technology is used to improve the outcome in every aspect of our lives. At the beginning of development, scientists made tools and pieces of stuff in order to enhance the speed of communication. The purpose of our research is to use modern technology to upgrade the lifestyle of human beings with the people who are struggling with obstacles. The machine interpretation of sign language has been conceivable yet in a restricted design, starting around 1977. At the point when an examination project effectively paired English letters from a console to ASL manual set letters which were reenacted on a mechanical hand. These innovations make an interpretation of sign language into a communicative language to communicate via gestures. The point of what is being looked for is now coming up. It’s already started to develop tools in order to make the communication procedure easier for people who can communicate with others through sign language. The objective of our endeavor is to provide a means of communication that facilitates interaction between those who possess normal hearing abilities and those who are deaf. The proposed system aims to identify indicators of deafness in individuals and use natural language processing (NLP) techniques to turn these indicators into a language that is readily understood, hence facilitating seamless communication between individuals with and without hearing impairments. The BDSL was used to enrich the dataset. In the event that an individual desires to use the model for a different language, it becomes required to make an update of the dataset. Our motto is - “Communications for everyone”.

Description

Cataloged from the PDF version of the thesis.
Includes bibliographical references (pages 32-33).
This thesis is submitted in partial fulfillment of the requirements for the degree of Master of Science in Computer Science, 2023.

Keywords

Deaf people, Camera vision, Real-time communication, Sign language, Mediapipe

Citation

Endorsement

Review

Supplemented By

Referenced By