A Comprehensive System for Real-Time Sign Language Interpretation Using Text and Speech Synthesis

A Comprehensive System for Real-Time Sign Language Interpretation Using Text and Speech Synthesis

 

 

Mr. Shashidhara H.V

Associate professor

Dept of Computer Science and Engg.

Malnad College of Engineering

Hassan, India

 

Mr. Jnanesh R

Dept of Computer Science and Engg.

Malnad College of Engineering

Hassan, India

 

 

Mr. Karthik V

Dept of Computer Science and Engg.

Malnad College of Engineering

Hassan, India

 

 

Mr. Kiran I S

Dept of Computer Science and Engg.

Malnad College of Engineering

Hassan, India

 

 

Mr. Hemanth I R

Dept of Computer Science and Engg.

Malnad College of Engineering

Hassan, India

 

 

Abstract:

Effective communication remains a challenge for individuals who rely on sign language as their primary mode of expression, especially in interactions with non-sign language users. This research explores an innovative system that converts sign language gestures into text and subsequently into synthesized speech, enabling seamless and inclusive communication. Leveraging advancements in computer vision, natural language processing (NLP), and speech synthesis, the proposed model captures real-time sign gestures, translates them into structured textual data, and outputs audible speech with high accuracy.

The study delves into key technologies, including machine learning algorithms for gesture recognition, dynamic language modelling for text interpretation, and scalable speech synthesis techniques for voice output. This paper aims to provide a comprehensive framework addressing the linguistic and technical complexities of sign-to-text-to-speech conversion, emphasizing its potential impact on accessibility and societal integration for the deaf and hard-of-hearing communities.