Using Speech to Assess Post-Traumatic Stress Disorder in Veterans
Post-Traumatic Stress Disorder (PTSD) is a serious problem for the military – it affects at least 30 percent of military service members who have spent time in war zones. Not only does PTSD affect veterans, it also impacts their families and communities. Moreover, about eight percent of the civilian population is likely to develop PTSD at some point in their lives—making it critical to detect and monitor in this group too.
Diagnosis is currently based largely on patient self-reporting during clinical interviews. There are few qualitative or objective measures available to help clinicians make a diagnosis and keep pace with growing needs. Correct diagnosis can be challenging due to factors such as distortions in memory and self-perception. In addition, clinical interviews require the patient to visit the clinician’s office and spend time being interviewed, which some patients may not be willing or able to do. There is a clear need for a more objective, time-efficient way to assess PTSD.
In collaboration with NYU Langone Medical Center, SRI International recently published a paper that examines the feasibility of using speech analytics for assessing PTSD in veterans. Based on prior research, there is evidence that speech is influenced by emotional and mental health state. Depressed people, for example, stereotypically speak with a flat affect or monotone. In this clinical study, SRI focused on the prosodic characteristics of speech, such as speaking rate, pitch, energy or intensity, and pause duration, as well as other acoustic features.
The ability to assess PTSD through speech has several advantages. Speech is natural, noninvasive, inexpensive, and can be obtained via phone for remote analysis. It may be used for triage purposes or to monitor treatment progress. Similar to body temperature, it may provide an indicator for the patient’s condition.
Initial results show promise for predicting a patient’s condition and provide evidence that speech can be used as an objective, useful indicator of PTSD status. SRI is continuing to analyze research subjects and investigate additional speech features, including lexical choices and disfluencies.
The goal of using speech for assessment extends beyond diagnosing PTSD. SRI’s Speech Technology and Research (STAR) Laboratory is focusing on speech analytics that can be relevant to a range of conditions, including depression, suicide risk and childhood trauma. Researchers Elizabeth Shriberg, Andreas Tsiartas, and Vikramjit Mitra have investigated several speech features that show promising results in each of these areas, suggesting that speech may be a valuable tool in the broad field of mental health. In the future, we plan to share additional insights on speech assessments that can bring potential benefits to people who suffer from these conditions.
This research and development project was conducted by SRI International and is made possible by a contract that was awarded and administered by the U.S. Army Medical Research & Acquisition Activity (USAMRAA) and the Telemedicine & Advanced Technology Research Center (TATRC), at Fort Detrick, MD, under contract: W81XWH-11-C-0004. The views, opinions and/or findings contained in this publication are those of the author(s) and do not necessarily reflect the views of the Department of Defense and should not be construed as an official Army position, policy, or decision unless so designated by other documentation.
Speech, Language, and Audio