Intellectual Property India Publishes Patent Application for 'Contextual Understanding In Voice-Based Question Answering Systems Using Deep Learning And Bert' Filed by Vellore Institute Of Technology

MUMBAI, India, Jan. 2 -- Intellectual Property India has published a patent application (202541123292 A) filed by Vellore Institute Of Technology, Vellore, Tamil Nadu, on Dec. 7, 2025, for 'contextual understanding in voice-based question answering systems using deep learning and bert.'

Inventor(s) include Dr. Ramesh C; Mr. Gugan S K; and Mr. Jayaa Suthersen A R.

The application for the patent was published on Jan. 2, under issue no. 01/2026.

According to the abstract released by the Intellectual Property India: "The present disclosure provides a contextual understanding system for voice-based question answering. The system includes a voice-based question answering module that processes YouTube video content by extracting audio, generating transcripts using automatic speech recognition, and generating contextually relevant answers using a transformer-based model. A PDF document processing module extracts textual content from PDF documents, segments content into chunks, and generates answers using large language models. An automatic speech recognition component converts voice-based user queries to text format. A transformer-based natural language processing model provides bidirectional contextual understanding of user queries and content. A vector database stores embeddings from processed content and enables semantic similarity searches for query matching. An enhanced multimodal PDF analysis module processes PDF documents containing textual and visual elements and generates responses using a large language model."

Disclaimer: Curated by HT Syndication.