MUMBAI, India, Oct. 12 -- Intellectual Property India has published a patent application (202641118885 A) filed by Rns Institute Of Technology; Niveditha G Y; Varun N C; Siva Vaishali B; Sahana Sabah; Divya Shree S; Janardhan Singh K; and Vishal Gowda R on October 05, 2026, for Signmate – An Ai-Driven Multimodal Communication System For Speech And Hearing Impaired.
Inventors include Niveditha G Y; Janardhan Singh K; Divya Shree S; Sahana Sabah; Siva Vaishali B; Varun N C; and Vishal Gowda R.
The application for the patent was published on October 09, 2026, under issue no. 41/2026.
Abstract: Abstract Communication with the people who do not know the sign language is a big problem for the hearing impaired people. Although several sign language recognition systems have been developed, many of them are limited to identifying individual gestures and do not provide support for complete communication. SignMate is a real-time Indian Sign Language (ISL) recognition system designed to address this issue using a standard webcam. The system captures hand movements from live video and extracts twenty-one hand landmark points through computer vision techniques. Two machine learning models are used for gesture recognition. For recognition of static gestures like alphabets and numbers, we use Support Vector Machine (SVM), and for dynamic word gestures which involves hand movement, a CNN-BiLSTM model is employed. The output that is recognised is converted to text and can be added to a sentence builder. This allows users to make meaningful sentences. The generated sentence can then be translated into a selected language and converted into speech by a text-to-speech module. The system combines gesture recognition, sentence formation, language translation and speech generation to enable better communication between sign language users and non-signers. Since the solution only works with a webcam and no special hardware is needed, it is an accessible communication aid for hearing-impaired people.
Disclaimer: Curated by HT Syndication.