Featured Image Credit: NurPhoto/Mike Kemp/Getty Images
How well can you say a complicated tongue twister after a heavy night of drinking?
Slurring your words is an unfortunate part of having a few too many, and soon your smartphone might be able to tell how drunk you are – by the sound of your voice.
A new Stanford University study tasked 18 participants with reading a tongue twister before drinking – which they then had to repeat once an hour for up to seven hours, after getting through an increasing amount of booze.
While doing so, smartphone sensors were used to record the study participants’ voices.
Advert
The aim was to explore the possibilities of turning our phones into smart breathalyzers, to avoid alcohol-related injuries and deaths in the future – and the results were impressive.
According to the World Health Organization, three million deaths occur every year globally as a result of harmful alcohol use, representing over 5% of all deaths worldwide.
And since just about everyone owns a smartphone these days, the breakthrough’s potential in preventing alcohol-related injuries and deaths could have a considerable impact on public health.
The result of the study? Smartphones could predict the participants’ intoxication with an astounding accuracy of 98%.
Associate professor of emergency medicine at Stanford, Brian Suffoletto, who led the research, said the accuracy of the software his team developed uses “cutting-edge advancements in signal processing, acoustic analysis, and machine learning”.
He added it could deliver “just-in-time interventions” to prevent injury and death resulting from motor vehicle or other accidents.
But Professor Suffoletto doesn’t necessarily want the tech to be used by periodically checking in with someone to see how drunk they are, and said that could “backfire by being annoying (at best) or by prompting drinking (at worst)”.
Instead, he sees it as being used passively – working in the background and letting someone know when they might need help.
“Timing is paramount when targeting the optimal moment for receptivity and the relevance of real-time support,” he said.
“For instance, as someone initiates drinking, a reminder of their consumption limits can be impactful. However, once they’re significantly intoxicated, the efficacy of such interventions diminishes.”
It’s a small study and more research needs to be done, but you never know, it might not be long before we see these health features in the likes of wearables like the Apple Watch – which already has a fall detection feature baked in.
Perhaps voice pattern sensors that help give health warnings on intoxication levels will be the next innovation to hit mainstream devices.
