MUMBAI, India, Sept. 22 -- Intellectual Property India has published a patent application (202611091694 A) filed by Thapar Institute Of Engineering & Technology on July 28, 2026, for An Edge-Optimized Intelligent Robotic Communication System For Real-Time Speech-To-Sign Translation.
Inventors include Dr. Ankit Soni; Dr. Mahesh Shamrao Chaudhari; Adamya Sharma; and Kartik Sindhi.
The application for the patent was published on September 18, 2026, under issue no. 38/2026.
Abstract: The present invention relates to an edge-optimized intelligent robotic communication system for real-time speech-to-sign translation, designed to facilitate communication for individuals with hearing and speech impairments. The system comprises an input module for receiving spoken language, an edge computing device configured to perform speech preprocessing and automatic speech recognition (ASR), and a Large Language Model (LLM). A CNN–LSTM-based gesture mapping module converts the processed textual information into corresponding sign language gesture identifiers, while an adaptive gesture sequencing module generates smooth and natural gesture sequences. A microcontroller receives gesture command signals and actuates a plurality of servo motors to operate a robotic hand for real-time execution of sign language gestures. The system further incorporates a feedback mechanism for improving translation accuracy and personalization through continuous learning. The edge-based architecture minimizes latency, enhances privacy, improves translation accuracy, and provides a scalable, energy-efficient, and context-aware assistive communication solution.
Disclaimer: Curated by HT Syndication.
