Year: 2025
Type: article
Abstract: The word-based Indian Sign Language (ISL) translator using sequence-to-sequence (Seq2Seq) model is a gated recurrent unit-based model, which is designed to establish communication for the deaf and hearing-impaired people in India with spoken language users. It utilizes OpenCV, MediaPipe for precise landmark recognition, and a Seq2Seq model built using TensorFlow to predict the actions. This combination allows the system to translate ISL gestures into text or speech in real time, overcoming the unique grammar and syntax challenges of ISL. Trained on an ISL dataset collected from students studying ISL, this model gives accurate and real-time translations of hand signs and gestures. This system demonstrates significant potential in bridging communication gaps and improving the lives of ISL users.
Authors: Fatima A Ansari, Vivek S Chouhan, Nishikant S Raut, Rehan F Sayyed, Rohit S Deshmukh
Language: English
Subfield: Human-Computer Interaction
Field: Computer Science
Domain: Physical Sciences
Open Access status: hybrid
Estimated APC paid:
OpenAlex ID: W4414013366