Scholarlink

<p>In the current scenario, where we can see young people struggling for their careers, they are even fighting a battle with their stress and tension. None of their work is done without stress to complete their task and compete with others. To overcome stress, one should have good emotional intelligence to cope with emotions and any upcoming stress. But at some point, due to lack of guidance, some people don’t know how to analyze the situations and how to handle them without taking the stress and end up with anxiety, depression, disappointment, suicide, heart attack, stroke etc. Due to the advancement of Human–Computer Interaction (HCI), medical science has leveled up to another peak. Machine Learning and Deep Learning played a major role in such interactions and predictions. Many applications have been developed in past years based on machine learning and deep learning. One of those applications is related to psychology and is still in research. These applications can be used for emotion and stress analysis among people, especially youngsters. Research in this field is being conducted using various verbal and non-verbal parameters. This paper addresses the research problem of improving emotion recognition accuracy and robustness to better analyze and manage stress. The primary objective is to develop an advanced Emotion Recognition System (ERS) that leverages deep learning algorithms to analyses both verbal and non-verbal cues—specifically, speech and body posture, including facial expressions. We have further integrated it with the Flask web framework to make an Emotion Recognition System that takes input in the form of video and audio to analyze Emotions and Stress. We have also compared our proposed ERS with existing ones and found that our ERS gives better results.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

A deep learning approach to analyse stress by using voice and body posture

  • 期刊论文

  • 发表日期：

    2025年2月19日

查看全文
  • Sumita Gupta,
  • Sapna Gambhir,
  • Mohit Gambhir,
  • Rana Majumdar,
  • Avinash K. Shrivastava,
  • Hoang Pham

摘要

In the current scenario, where we can see young people struggling for their careers, they are even fighting a battle with their stress and tension. None of their work is done without stress to complete their task and compete with others. To overcome stress, one should have good emotional intelligence to cope with emotions and any upcoming stress. But at some point, due to lack of guidance, some people don’t know how to analyze the situations and how to handle them without taking the stress and end up with anxiety, depression, disappointment, suicide, heart attack, stroke etc. Due to the advancement of Human–Computer Interaction (HCI), medical science has leveled up to another peak. Machine Learning and Deep Learning played a major role in such interactions and predictions. Many applications have been developed in past years based on machine learning and deep learning. One of those applications is related to psychology and is still in research. These applications can be used for emotion and stress analysis among people, especially youngsters. Research in this field is being conducted using various verbal and non-verbal parameters. This paper addresses the research problem of improving emotion recognition accuracy and robustness to better analyze and manage stress. The primary objective is to develop an advanced Emotion Recognition System (ERS) that leverages deep learning algorithms to analyses both verbal and non-verbal cues—specifically, speech and body posture, including facial expressions. We have further integrated it with the Flask web framework to make an Emotion Recognition System that takes input in the form of video and audio to analyze Emotions and Stress. We have also compared our proposed ERS with existing ones and found that our ERS gives better results.