MUMBAI, India, March 13 -- Intellectual Property India has published a patent application (202641024380 A) filed by Vellore Institute Of Technology, Vellore, Tamil Nadu, on March 1, for 'a real-time interactive smart reader system for facilitating assisted digital reading and comprehension.'

Inventor(s) include Parvathi R; and N. Janani.

The application for the patent was published on March 13, under issue no. 11/2026.

According to the abstract released by the Intellectual Property India: "The present invention relates to a real-time interactive smart reader system for 5 assisted digital reading and comprehension. the system is configured to receive digital documents such as PDF files, text articles, and plain text files, extract machine-readable content, normalize the text, and segment it into structured lines and tokenized sentences while preserving the original order. a multistage natural language processing (NLP) intelligence framework performs linguistic 10 preprocessing, generates extractive summaries using a graph-based sentence ranking model such as TextRank, and provides multilingual translation for user- selected segments. the system further includes an interactive reading module enabling user-defined line selection, text-to-speech conversion, and synchronized audio playback with dynamic line-level and word-level highlighting, along with 15 real-time controls. a context-aware vocabulary guidance engine retrieves lexical metadata including meanings, synonyms, antonyms, pronunciation, and example usage using resources such as WordNet without interrupting reading flow. additionally, a simplified language rewriting module reduces linguistic complexity through rule-based lexical substitution while maintaining semantic 20 integrity and formatting. an orchestration module ensures low-latency temporal alignment between speech and highlighted text. the system outputs synchronized multimodal content including audio, translated text, simplified text, and vocabulary overlays, thereby improving accessibility, comprehension, and retention without requiring user-specific model training datasets."

Disclaimer: Curated by HT Syndication.