MUMBAI, India, July 24 -- Intellectual Property India has published a patent application (202641086085 A) filed by Vallurupalli Nageswara Rao Vignana Jyothi Institute Of Engineering Technology on July 14, 2026, for Signosphere: An Intelligent Gesture-To-Speech Communication System Using Vision Transformers And Real-Time Dynamic Sentence Generation.
Inventors include Dr. P. Bharath Kumar Chowdary; Mr. Peddarapu Rama Krishna; G. Laxmi Deepthi; K. Hari Priya; K. Bhagya Rekha; and Ch. Sandhya Rani.
The application for the patent was published on July 17, 2026, under issue no. 29/2026.
Abstract: ABSTRACT [0015] According to recent studies, individuals with speech disabilities frequently encounter significant communication barriers that hinder their ability to interact effectively with society. Those affected often experience social isolation and face challenges in accessing education and employment opportunities, impacting their mental health, well-being, career prospects, and overall personal development. It is crucial to address these barriers by creating and implementing innovative and inclusive communication tools to improve their quality of life and promote equal opportunities. Present solutions lack the ability to generate grammatically correct, contextually meaningful sentences and are limited by hardware requirements. The proposed system, Signosphere, aims to overcome these limitations by integrating technologies such as Kotlin (Android Studio), Python-based Vision Transformer (ViT) models, Mediapipe, OpenCV, and Google Text- to-Speech (gTTS). Through this integration, the system seeks to achieve accurate real-time gesture recognition, dynamic sentence formation, multi- language support, and efficient performance even on low-resource devices. This solution aspires to enhance inclusivity and ease communication for people with speech impairments, reducing barriers and fostering greater social integration.
Disclaimer: Curated by HT Syndication.
