Smartphones are an essential part of our daily lives, and now they may also be a valuable tool in detecting alcohol levels that are incompatible with driving. A study from Stanford University has revealed that an artificial intelligence model can analyze voice patterns with a 98% accuracy in detecting drunkenness.
The study involved giving participants a calculated amount of alcohol based on their weight and then asking them to recite tongue twisters before and after drinking. The AI model analyzed their voice patterns to predict alcohol poisoning, and the results were compared to breath alcohol tests.
The high degree of precision in this method has surprised even the researchers involved in the study. One of the main advantages of using smartphones for this purpose is that almost everyone has their cell phone on hand when they go out, making it the best tool to turn to in case of doubt.
However, researchers are aware that voice analysis alone may not be enough to determine alcohol intoxication, and more research with a larger sample size and a greater variety of people is needed to confirm the validity of voice patterns in detecting drunkenness. Not only that, but using a combination of variables such as voice analysis, walking patterns, and behavior when sending text messages may lead to more precise results.
While there are already applications to estimate blood alcohol levels, they are not always reliable, and the use of a conventional breathalyzer is still recommended. In fact, the DGT in Spain already obliges drivers of passenger transport vehicles to carry a breathalyzer in the car connected to the starting system to ensure that if they exceed the permitted limit, the car cannot be started.
In conclusion, while the use of smartphones for detecting alcohol intoxication shows promise, more research is needed to develop reliable software that can be used in moments of doubt. In the meantime, it is advisable to continue using traditional methods such as breathalyzers to ensure safe driving practices.