It turns out the human voice isn’t just for ordering replacement detergent from your Amazon Echo or sending a text message hands-free while driving. Voice health functionality could be coming soon to the big names in voice-activated technology. So, are the days of false-alarm visits to the doctor and the late detection of ailments numbered?
Researchers from New York University’s Langone Medical Center have conducted some new research, reported Forbes, that points to this possibility. Collaborative research from a number of institutions is starting to show short audio clips of the human voice may be used to diagnose a variety of diseases, psychiatric disorders and other health conditions — in some cases, potentially sooner than current diagnostic processes allow.
Nonprofit research center SRI International, for example, reported early results of having identified some 30 characteristics out 40,000 total features and other features that can be used to diagnose PTSD, said Forbes, with 77 percent accuracy.
The Mayo Clinic has linked some 13 vocal features in common from patients with heart disease. The MIT Technology Review wrote that Mayo also has plans to conduct a similar study in China to determine if these features translate into other languages.
Jim Harper, CEO of Sonde Health in Boston, which is working on voice tests to monitor for postpartum depression, Parkinson’s and dementia, was quoted as saying: “We’re trying to make this ubiquitous and universal by engineering a technology that allows our software to operate on mobile phones and a range of other voice-enabled devices.”
So, could Siri, Amazon’s Alexa, Cortana, Google Home and others be just a few years away from diagnosing users from their smartphones or while sitting on their kitchen counter? Potentially — if the research pans out, further studies find consistent markers within an individual or across different languages and patients can’t find a way to easily trick or manipulate the voice-diagnostic systems.
But the problem isn’t just in whether the human voice can be used as an accurate diagnostic tool. It’s one thing if the voice technology finds itself in physicians offices and emergency rooms; it’s an entirely different story if the capability extends to personal devices. If and once a system is developed, a host of privacy, security and safety concerns arise around consumers using IoT devices for medical diagnoses.
It’s no secret that the IoT has some work to do on the security end. Likewise, consumers are currently eyeing voice-activated devices with suspicion. Despite the worries, most American consumers are more than happy (on average) to bring voice-activated tech into their homes.
A recent smart home marketplace survey conducted by Coldwell Banker and Vivint Smart Home found that some 72 percent of Americans who have smart home products want voice control, and further that some 48 percent of Americans already own devices with voice-activated functionality.
The largest use case of voice-activated products in American homes thus far is to control entertainment, at 57 percent. Controlling smart lighting, security products and shopping all saw 33 percent.
While America is busy bonding with Alexa, Google Home or Cortana-enabled devices, its neighbor to the north is still waiting on a full Canadian release of a major voice-activated AI device.
The Toronto Star suggested a few reasons for this, including the significantly smaller market making Canada a lower priority, as well as the need to create bilingual packaging and French-language features for Google’s and Amazon’s offerings. The only thing tougher than creating a natural language processor is creating additional models in different languages.
While it’s certainly possible for Canadian consumers to buy a Google Home or an Amazon Echo stateside and bring it back home, the functionality isn’t quite the same once a voice-enabled device crosses the border. For example, while most of Echo’s features work north of the border, voice ordering requires a U.S. credit card and address.
Likewise, location services aren’t there yet. The Star said Canadian users have to ask for the weather in their town specifically — “What’s the weather in Moose Jaw, Saskatchewan?” as opposed to the more general “What’s the weather?”
Versions of Echo and Echo Dot were released in the U.K. and Germany back in September, so it’s not like Amazon is holding out on international or multilingual releases. So, while some day Canadian consumers will be able to order fresh batteries from their Echo or control their burgeoning smart home with Google Home, it may take some time.