MUMBAI, India, June 26 -- Intellectual Property India has published a patent application (202641071966 A) filed by Dr. M. Chithra; Dr. M. Arthi; Dr; R. Geetha; Dr. Sheeba Manodh; Dr. Vjayanthi; N. Suguna; Mrs. J. Rajkamal Petro; P. Tamilarasan; Dr. Vijayakumar Selvaraj; Dr. Sheik Hameed Nagoor Gani; and Dr. N. Gopinath on June 10, 2026, for An Ai Ml Based Emotion Aware English Text Generation Using Neural Language Models.
Inventors include Dr. M. Chithra; Dr. M. Arthi; Dr; R. Geetha; Dr. Sheeba Manodh; Dr Vjayanthi; N. Suguna; Mrs. J. Rajkamal Petro; P. Tamilarasan; Dr. Vijayakumar Selvaraj; Dr. Sheik Hameed Nagoor Gani; and Dr. N. Gopinath.
The application for the patent was published on June 19, 2026, under issue no. 25/2026.
Abstract: AN Al ML BASED EMOTION-AWARE ENGLISH TEXT GENERATION USING NEURAL LANGUAGE MODELS 10-Jun-2026/86143/202641071966/Form 2(Title Page) ABSTRACT The rapid advancement of Artificial Intelligence (Al) and Natural Language Processing (NLP) has enabled machines to generate human-like text for various applications. However, most conventional text generation systems primarily focus on grammatical correctness and contextual relevance while lacking the ability to understand and express human emotions effectively. The proposed invention, “Emotion-Aware English Text Generation Using Neural Language Models,” introduces an intelligent text generation framework capable of producing emotionally adaptive and contextually coherent English text. The system integrates advanced neural language models with emotion recognition and sentiment analysis mechanisms to identify emotional cues from user inputs, conversational history, or external contextual data. The proposed framework employs deep learning architectures, including transformer-based neural networks, attention mechanisms, and emotion embedding layers, to capture emotional states such as happiness, sadness, anger, fear, surprise, and neutrality. Based on the detected emotional context, the model dynamically adjusts vocabulary selection, sentence structure, tone, and style to generate text that aligns with the intended emotional expression. The system further incorporates reinforcement learning and feedback-driven optimization techniques to improve emotional accuracy, linguistic quality, and user engagement over time. The invention can be applied in diverse domains such as intelligent chatbots, virtual assistants, customer support systems, educational platforms, mental health applications, content creation tools, and social communication systems. Experimental evaluations demonstrate that the proposed emotion-aware text generation model significantly enhances emotional relevance, readability, contextual consistency, and user satisfaction compared to traditional language generation approaches. By bridging the gap between human emotional communication and machine-generated language, the proposed system offers a robust, scalable, and adaptive solution for next-generation emotionally intelligent text generation applications.
Disclaimer: Curated by HT Syndication.
