Can our voices reveal how we feel under stress or emotional load? This systematic review by our team at the University of Bern and the University Hospital of Bern explores how speech—specifically acoustic features like pitch, intensity, and speech rate —can serve as a non-invasive method to assess negative emotions, stress, and cognitive load.
Reviewing 38 articles across four major databases, we found that prosodic features (such as fundamental frequency) were the most informative for identifying negative emotions, stress, and cognitive load. Specifically, anger was associated with elevated fundamental frequency (F0), increased speech volume, and faster speech rate. Stress was associated with increased F0 and intensity, and reduced speech duration. Cognitive load was linked to increased F0 and intensity, although the results for F0 were overall less clear than those for negative emotions and stress. The evidence for emotions like fear and anxiety remains mixed, highlighting the need for more targeted research.
This work offers key insights for both researchers and practitioners interested in emotion recognition, affective computing, and stress monitoring in real-world settings.
Link to the full article, published in PLOS One: https://journals.plos.org/plosone/article?id=10.1371/journal.pone.0328833
