Detection of Stress by Voice & Textual Information Analysis

The goal of our project is to analyze stress in human speech. Much work has already been done on this area. However, for our project, we have set the goal to identify stress from a speech by analyzing the voice of an individual and to convert that same speech into textual information and try to capture the stress from it. We aim to combine these two approaches and their results to give a more accurate approach to stress analysis in human speech. Our project has two parts: For the first part we aim to detect and analyze stress from voice, an approach better known as Voice Stress Analysis (VSA). For the second part, we aim to convert the same speech into textual format and then capture the emotions related to “Stress ” from that textual information. Background Information: Part I: Stress in human speech can be detected by various methods known as Voice Stress Analysis (VSA). Voice stress analysis originated from the concept that when a person is under stress, micro-muscle tremors (MMT) occur in the muscles that make up the vocal tract which are transmitted through the speech. An MMT is a low amplitude oscillation of the reflex mechanism controlling the length and tension of a stretched muscle caused by the finit