MUMBAI, India, Nov. 2 -- Intellectual Property India has published a patent application (202641124064 A) filed by Koneru Lakshmaiah Education Foundation on October 16, 2026, for Ai-Enabled Offline Wearable System And Method For Real-Time Bidirectional Sign Language Recognition, Translation And Communication.
Inventors include Sunkara Eswar; Sai Pradhun Gudipudi; and S. Shanmugan.
The application for the patent was published on October 30, 2026, under issue no. 44/2026.
Abstract: An offline wearable system (100) enables two-way communication between a sign-language user and a person who does not sign. A glove carries flex sensors (202) along the fingers and an inertial measurement unit (204); a wrist module (206) houses a microcontroller (122), display, speaker, vibration motor, microphone and battery. The microcontroller normalises the sensor signals with per-user calibration, segments individual signs by motion energy, classifies them with a quantised neural network, and outputs confident results as text and speech while asking for uncertain signs to be repeated. Replies from the other person, spoken or entered by buttons, are recognised by an offline keyword recogniser (144), mapped to messages and presented to the wearer as text, sign glyphs and distinct haptic vibration codes. All processing is performed in the device, without a smartphone, computer or network.
Disclaimer: Curated by HT Syndication.
