MUMBAI, India, Sept. 28 -- Intellectual Property India has published a patent application (202641112147 A) filed by Mlr Institute Of Technology; and Marri Laxman Reddy Institute Of Technology And Management on September 18, 2026, for Machine Learning Based User – Friendly Multilingual Voice Message Translator To Sign Language.
Inventors include Mrs. K. Pushpa Rani; Ms. Chinthaguddi Sravani; Mr. S. Uday Kiran Reddy; and Mr. S. Sravanth.
The application for the patent was published on September 25, 2026, under issue no. 39/2026.
Abstract: In this invention, a machine learning-based multilingual voice message translator to sign language is provided for enabling real-time communication for individuals with hearing or speech impairments. The system comprises a voice input module configured to capture spoken audio through mobile, web, or embedded applications and transmit the audio to a cloud-based processing platform. A deep learning engine, including convolutional neural network (CNN) and long short-term memory (LSTM) architectures, is configured to perform automatic language identification and automatic speech recognition (ASR) to convert speech into textual data, and further map the textual data into corresponding sign language representations using a structured multilingual sign database. A sign generation module is configured to produce synchronized hand gestures, facial expressions, and body postures, which are rendered through animated avatars or visual symbols for intuitive comprehension. The system further incorporates data preprocessing including cleaning, augmentation, feature extraction, and iterative model training to enhance accuracy and reliability across multiple languages. Cloud deployment enables scalable real-time processing, secure storage, and efficient handling of large datasets. The invention thereby facilitates accessible, inclusive, and effective communication across educational, healthcare, customer service, and social environments. 5 Claims & 1 Figure.
Disclaimer: Curated by HT Syndication.
