Let's Get Vocal About Disease

MP3 WAV

Alexa - tell me, do I have COVID-nineteen? Siri - am I developing dementia? Well, I can tell you that, Norbert! Ha! The thing is Dave, this is not a stretch. For the past decade, biotech companies have been developing voice analysis technology that can detect disease. And it's almost here! It's like I'm living in a Star Trek episode watching this crazy innovation move into reality.

The technology uses artificial intelligence and machine learning to identify vocal biomarkers. These bio-signatures are present in dementia, depression, autism, and heart disease.

When your body responds to disease, several systems become involved. For example, in early COVID infection, your lungs and larynx are both affected. In dementia or Parkinson's disease, speed of speech is changed and there's halting pauses as well as low volume.

Researchers fed voice recordings from thousands of patients into computer algorithms that could analyze voices. AI and machine learning check for changes in speech patterns, intensity, tone and other features that become bio-signatures of the disease's impact on your speech. The accuracy in a number of studies already shows a detection rate of at least ninety percent.

Alexa won't have the final word though. A doctor would have to make the official diagnosis, but virtual medicine may hit warp speed once this technology beams into our homes.

What could be next?!

More Information

Alexa, do I have COVID-19?
Researchers are exploring ways to use people's voices to diagnose coronavirus infections, dementia, depression and much more...

Voice Signal Characteristics Are Independently Associated With Coronary Artery Disease
Voice signal analysis is an emerging noninvasive diagnostic tool. The current study tested the hypothesis that patient voice signal characteristics are associated with the presence of coronary artery disease (CAD)...

Investigation of Speech Landmark Patterns for Depression Detection
This article proposes two feature sets associated with speech articulation events based on counts and durations of sequential landmark groups or n-grams...