MUMBAI, India, June 22 -- Intellectual Property India has published a patent application (202641050122 A) filed by Dr. N. Bharathi on April 20, 2026, for Healing Based Iot Monitoring System Using Hardware Redundancy And Software Intelligence.
Inventor includes Dr. N. Bharathi.
The application for the patent was published on June 12, 2026, under issue no. 24/2026.
Abstract: ABSTRACT The present invention discloses an Artificial Intelligence-based system and method for multimodal stress detection using speech signals and natural language processing techniques. The system integrates audio feature extraction and sentiment analysis to accurately identify stress levels in human speech, enabling intelligent emotional state assessment in real-time applications. The proposed system processes speech input in the form of audio files and extracts relevant acoustic features such as Mel-Frequency Cepstral Coefficients (MFCC), Zero Crossing Rate (ZCR), and Root Mean Square (RMS) energy, which capture the tonal and intensity characteristics of the voice. In parallel, a speech-to-text module converts-th~ into textual data,. which- is further analyzed--using-a-sentiment analysis model to derive emotional polarity scores including negative, neutral, positive, and compound sentiments. A machine learning-based classification model, specifically a Random Forest classifier, is employed to combine both audio and textual features into a unified representation for robust stress prediction. The system classifies input speech into stressed and non-stressed categories based on learned patterns from the dataset. The use of multimodal inputs enhances prediction accuracy and reliability compared to singlemodality approaches. The system further incorporates an interactive web-based interface developed using Gradio, enabling users to upload audio samples and obtain real-time stress predictions along with transcribed text outputs. The model is trained on a structured emotional speech dataset and supports scalable deployment for practical applications. The invention provides a cost-effective and efficient solution for stress detection in domains such as mental health monitoring, workplace wellness, human-computer interaction, and telehealth systems, offering a non-invasive and automated approach to emotional analysis.
Disclaimer: Curated by HT Syndication.
