Non-verbal communication translator for speech impaired using CNN

Disability makes communication difficult. This paper’s tasks study the identification of ISL characters (ISL). Due of its complexity and considerable hand gestures, this activity has a huge societal effect, yet it’s challenging. They need a mediator who can translate sign language into writing. Norm...

Full description

Saved in:
Bibliographic Details
Main Authors: Ramachandro, M., Reddy Gunna, Shivani, Saahithi, Doma, Manvith, Gundeti Sai
Format: Conference Proceeding
Language:English
Subjects:
Alphabets
People with disabilities
Task complexity
Translators
Verbal communication
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:Disability makes communication difficult. This paper’s tasks study the identification of ISL characters (ISL). Due of its complexity and considerable hand gestures, this activity has a huge societal effect, yet it’s challenging. They need a mediator who can translate sign language into writing. Normal people can’t fully comprehend disabled people’s gestures, thus they require a translator. With a sign-language translator, communication is two-way. We can’t always look for a translator. We need a platform that identifies alphabets and numbers as text. Hearing-impaired people employ nonverbal sign language. Webcams capture hand motions. We used CNN classifier. We obtain 90-95 percent accuracy after training our model using the Kaggle Indian Sign Language hand dataset, which has 1200 photos for each alphabet and number.
ISSN:0094-243X
1551-7616
DOI:10.1063/5.0199401