Stress measurement using speech: Recent advancements, validation issues, and ethical and privacy considerations

Abstract

Life stress is a well-established risk factor for a variety ofmental and physical health problems, including anxiety disorders, depression, chronic pain, heart disease, asthma, autoimmune diseases, and neurodegenerative disorders.The purpose of this article is to describe emerging approaches for assessing stress using speech, which we do by reviewing the methodological advantages ofthese digital health tools, and thevalidation, ethical, and privacy issues raised bythese technologies.As we describe, itis now possible to assess stress via the speech signalusing smartphones and smart speakers that employ software programs and artificial intelligence to analyze several features of speech and speech acoustics, including pitch, jitter, energy, rate, and length and number of pauses.Because these digital devices are ubiquitous, we can nowassess individuals’ stress levels in real time in almost any natural environment in which people speak.These technologies thus have great potential for advancing digital health initiatives that involve continuously monitoring changes in psychosocial functioning and disease risk over time. However, speech-based indices of stress have yet to bewell-validated against stressbiomarkers (e.g., cortisol, cytokines) that predict disease risk. In addition, acquiringspeech samplesraises the possibility thatconversations intended to be private could one day bemade public; moreover, obtaining real-time psychosocial risk information prompts ethical questions regarding how these data should be used for medical, commercial, and personal purposes. Althoughassessing stress using speech has enormous potential, there are critical validation, privacy, and ethical issues that must be addressed.