Related Experiment Video
Updated: Oct 10, 2025
05:48
Author Spotlight: Investigating the Impact of Emotional Prosodies on Voice Recognition and Perception
Published on: August 9, 2024
1.7K
COVID-19: Affect recognition through voice analysis during the winter lockdown in Scotland
Summary
This study explored using voice analysis via home devices to remotely assess emotions during COVID-19 lockdowns. Machine learning models showed feasibility in detecting psychological wellbeing changes through acoustic features.
Area of Science:
- Psychology
- Computer Science
- Health Informatics
Background:
- The COVID-19 pandemic imposed lifestyle restrictions impacting psychological wellbeing.
- Prevalence of depression and anxiety symptoms in Scotland highlights the need for accessible mental health monitoring.
Purpose of the Study:
- To investigate social signal processing for remote emotion assessment.
- To develop and evaluate a machine learning method for affect recognition using voice recordings.
Main Methods:
- A machine learning model was developed for affect recognition.
- Acoustic features were extracted from voice recordings collected via home and mobile devices during the COVID-19 winter lockdown in Scotland.
- Random Forest and Decision Trees were used to predict arousal and valence.
Main Results:
- The Random Forest model achieved a concordance correlation coefficient of 0.4230 for arousal.
- The Decision Trees model achieved a concordance correlation coefficient of 0.3354 for valence.
- The study demonstrated the feasibility of remote, automated, and large-scale psychological wellbeing monitoring.
Conclusions:
- Remote emotion recognition using voice analysis from everyday devices is feasible.
- This technology can aid in early detection of mental health difficulties like depression and anxiety.
- Automated monitoring supports timely intervention, potentially preventing disability.
