Smart phones and speakers detecting in our speech how much alcohol we have drunk

2023-11-10 23:30:32

Thanks to sensors in smartphones and smart speakers, these devices could determine with remarkable precision and reliability how much alcohol the person or people with whom they live has consumed. To do this, they would be guided by the changes in the way of speaking that these people exhibit, compared to when they are sober.

This has been determined in a series of experiments by scientists from Stanford University in the United States and Toronto University in Canada.

This team, led by Brian Suffoletto of Stanford University, conducted a small study on 18 adults ages 21 and older. Participants were given a dose of alcohol based on their weight and were randomly assigned a series of tongue twisters: one before drinking and one every hour for up to seven hours after drinking.

Participants were asked to read the tongue twister aloud, and a smartphone was placed on a table 30 to 60 centimeters away to record their voices. The researchers also measured their breath alcohol concentration at the start of the study and every 30 minutes for up to seven hours later.

These intelligent devices successfully analyzed the speech of the assigned people, detecting changes that revealed their alcohol consumption.

When compared to the results of a breathalyzer test, changes in participants’ voice patterns as the experiment progressed predicted alcohol intoxication with 98% accuracy.

Smartphones could detect alcohol poisoning by analyzing voice patterns. (Photo: Journal of Studies on Alcohol and Drugs)

This raises the possibility that in the future these devices will be trained to make decisions on their own and protect their users when they detect that they are in some kind of danger due to having consumed too much alcohol.

Suffoletto and his colleagues present the technical details of the results of their experiments in the academic journal Journal of Studies on Alcohol and Drugs, under the title “Detection of Alcohol Intoxication Using Voice Features: A Controlled Laboratory Study.” (Source: NCYT from Amazings)