Study Confirms Sonde Health's Voice-Based Mental Fitness Solution Accurately Identifies Individuals with Elevated Mental Health Symptoms

BOSTON - March 11, 2024 - A new study from Sonde Health has validated the company's mental fitness vocal biomarker (MFVB) platform’s ability to reliably distinguish individuals with elevated mental health symptoms. The four-week cohort study revealed a statistically significant correlation between voice-based identification of increased or decreased mental health risk with the results of the M3 Checklist, a clinically validated mental health assessment. 

The research, published in the peer-reviewed journal Frontiers in Psychiatry, highlights the potential of vocal biomarkers and Sonde’s technology specifically to provide objective data that can complement clinical care and improve self-monitoring for conditions like depression, stress- and trauma-related conditions, and anxiety.

Over the course of four weeks, study participants used Sonde’s Mental Fitness smartphone app to record their thoughts and feelings as a 30-second voice journal. The MFVB monitoring tool analyzed those free speech recordings for eight acoustic features relevant to mental health — jitter and shimmer, pitch variability, energy variability, vowel space, phonation duration, speech rate, and pause duration — and calculated a real-time MFVB score ranging from 0-100. Scores of 80 to 100 were labeled “Excellent,” 70-79 were “Good,” and 0-69 were categorized as “Pay attention.” MFVB scores were then cross-referenced against the results of participants’ M3 Checklist.

Participants were twice as likely to report elevated mental health symptoms if their MFVB scores remained in the “Pay Attention” range versus in the “Excellent” range over a period of two weeks. This effect was even more pronounced for participants who engaged with the tool more frequently. For those who used it 5-6 times per week, the likelihood of elevated symptoms via the M3 Checklist varied by a factor of 8.5 across this range of MFVB scores.

"This study further validates our voice-based health tracking platform as an objective indicator of mental well-being," said Erik Larsen, Senior Vice President of Clinical Development & Customer Success at Sonde Health. "The results show vocal biomarkers can provide meaningful insights into mental health in a preventive, scalable way. We believe the MFVB can foster stronger awareness about individuals’ mental well-being, thereby encouraging them to cultivate healthy habits and proactively mitigate their mental health risks."

Participants in the study exhibited positive engagement and reported favorable experiences with the MFVB tool. Approximately 40% changed their behavior or lifestyle in some way, and 30% perceived benefits to their well-being. Seventy-two percent of participants wanted to continue using the MFVB app to track how they’re doing in the future.

"The ability to collect mental health data from patients between clinic visits could transform how we monitor symptoms and optimize treatment plans," said Lindsey Venesky, Ph.D., Licensed Psychologist and Clinical Director at the Cognitive Behavior Institute (CBI), which collaborated on the study. “Voice-based health tracking technology can provide accurate insights into a client’s mental health status over time and can do so seamlessly and unobtrusively, with little added effort for clients.” 

The study enrolled 104 outpatient psychiatric participants with at least one clinician-verified symptom of depression from the Cognitive Behavior Institute in Pittsburgh, Pennsylvania. Participants were evaluated and encouraged to interact with the MFVB app as frequently as desired over a four-week period between May and August 2023.

Sonde’s MFVB technology has already been integrated into consumer health monitoring apps, life insurance health apps, consumer audio wearables, and automobiles. 

About Sonde Health

Sonde Health is the global leader in voice-based health tracking and data insights. Sonde’s vocal biomarker API/SDK serves enterprise apps and devices spanning consumer wellness to population health. Leveraging a best-in-class voice data set and clinical research with over 1.2 million samples from 85,000+ individuals on four continents, Sonde uses advanced audio signal processing, speech science, and AI/machine learning to sense and analyze subtle vocal changes due to changes in a person’s physiology to provide key insights into health and well-being. www.sondehealth.com