Smartphone app detects hypertension from sound of voice

A running joke for millennials and Gen Zers is feeling anxious to make phone calls because they’re used to communicating by text only, but a new breakthrough might actually help us lower our blood pressure by speaking on the phone.

Scientists at Klick Labs have developed an AI system that can identify signs of hypertension just by analyzing someone’s voice as they speak into their smartphone, potentially transforming how we screen for high blood pressure.

One in three adults worldwide has high blood pressure, yet many don’t know it. A recent study from JAMA confirms that in the United States, over half of the adults with uncontrolled hypertension are unaware they have this condition. This “silent killer” often goes undetected until it causes serious health problems. Traditional blood pressure measurements require specific equipment and technical know-how, making regular monitoring a challenge—especially in areas with limited healthcare access.

Klick Labs’ findings, published in IEEE Access, demonstrate how AI can detect subtle vocal patterns linked to high blood pressure. The team trained their AI models by analyzing speech features in the voice recordings of 267 adults. The system picked up on minute variations in voice characteristics that humans can’t hear—differences that turned out to be reliable indicators of hypertension.

“By leveraging various classifiers and establishing gender-based predictive models, we discovered a more accessible way to detect hypertension, which we hope will lead to earlier intervention for this widespread global health issue. Hypertension can lead to a number of complications, from heart attacks and kidney problems to dementia,” said Yan Fossat, senior vice president of Klick Labs and principal investigator of the study.

The technology works by breaking down voice recordings into their component parts, examining features like frequency, amplitude, and rhythm. These vocal fingerprints are then processed through machine learning algorithms that can spot patterns associated with high blood pressure. The system proved particularly effective when analyzing male voices, achieving an impressive 89% accuracy rate.

The AI system examines multiple aspects of speech patterns simultaneously. It analyzes the fundamental frequency of a person’s voice, studies how their vocal cords vibrate, and evaluates subtle changes in speech rhythm and energy. These measurements create a vocal profile that can indicate the presence of hypertension.

This breakthrough has the potential to significantly improve healthcare accessibility. Traditional blood pressure measurements require a visit to a healthcare provider, specific equipment, and proper technique. Voice analysis could offer a convenient, cost-effective screening tool that works through any phone or computer. This could be particularly valuable for people who lack easy access to healthcare facilities or those who need frequent monitoring.

Klick Labs, known already for its innovative work in healthcare technology, began exploring vocal biomarkers after recognizing the need for more accessible diagnostic tools. Their research team saw the potential in voice analysis as a non-invasive way to detect various health conditions. The success of this project builds on years of research into how physical health conditions affect vocal patterns.

“Voice technology has the potential to exponentially transform healthcare, making it more accessible and affordable, especially for large, underserved populations,” said Jaycee Kaufman, Klick Labs research scientist and co-author of the study. “Our ongoing research increasingly demonstrates the significant promise of vocal biomarkers in detecting hypertension, diabetes, and a growing list of other health conditions.”

Beyond hypertension detection, as AI technology continues to advance, voice analysis could become a powerful tool for early detection of other medical issues, including neurological conditions and mental health disorders. Voice technology powered by AI could revolutionize how we approach healthcare monitoring and diagnosis, ultimately leading to regular health screenings as convenient for patients as speaking into their smartphone from the comfort of their homes.

Any advancement that makes healthcare monitoring more proactive than reactive is worth celebrating, as it can lead to improved patient outcomes. Rather than waiting for symptoms to appear, patients could have their health status monitored continuously through everyday conversations. This would create a sort of early warning system that prompts them to seek medical attention before a condition becomes severe. 

According to the study in JAMA, this technology could actually have the greatest impact on younger adults. This group (ages 20–40) has more limited engagement with the health care system than older adults, yet is more likely to experience life events, such as pregnancy, that would require personalized advice on blood pressure management. The study concludes that “Effective management strategies and efforts are needed to increase hypertension awareness among young adults, especially young females,” including “assisting patients with obtaining validated self-measured blood pressure monitors.” Klick Labs’ AI system would be even easier for young adults to integrate into their daily lives over traditional blood pressure testing equipment, even self-measuring monitors.

Klick Labs’ medical breakthrough represents a shift in how we think about diagnosis. The most effective solutions may not require new tests or invasive procedures but instead rely on innovative ways to interpret signals our bodies send naturally. As this technology evolves, it could help millions of people take control of their health through something as natural as speaking. We might soon see a world where our voices become a key to understanding our overall health.