MUMBAI, India, July 24 -- Intellectual Property India has published a patent application (202641085096 A) filed by Sasi Institute Of Technology & Engineering; P. C. S. Nagendra Setty; T. Lakshmanarao; G. Prasanth Kumar; and P. V. V. Satya Eswar Rao on July 11, 2026, for A Noise-Robust And Speaker-Independent Speech Emotion Recognition System For Human–computer Interaction.
Inventors include P. C. S. Nagendra Setty; T. Lakshmanarao; G. Prasanth Kumar; and P. V. V. Satya Eswar Rao.
The application for the patent was published on July 17, 2026, under issue no. 29/2026.
Abstract: This invention presents a comprehensive review of Speech Emotion Recognition (SER) systems that enable machines to identify human emotions from speech signals. The study analyzes research published between 2012 and 2025, focusing on feature extraction methods, machine learning algorithms, deep learning architectures, datasets, and classification techniques. Existing SER systems face challenges such as noise, speaker variability, language differences, and poor cross-dataset performance. The review highlights the effectiveness of advanced deep learning models, particularly Transformer-based architectures and attention mechanisms, in improving emotion recognition accuracy. Techniques such as domain adaptation and cross-corpus learning are also examined. The findings identify current limitations and propose future directions for developing robust, accurate, and real-world SER systems capable of recognizing emotions across diverse languages, speakers, and environments.
Disclaimer: Curated by HT Syndication.
