Smartphones may be able to detect how drunk someone is, based on their voice

Smartphones may be able to detect how drunk someone is, based on their voice

Smartphones may be able to detect how drunk someone is? Yes by just their voice. Great innovation right?

A short study by researchers at Stanford University and the University of Toronto suggests that cellphones could be able to tell someone’s state of drunkenness based on changes in voice patterns.

Smartphone sensors were used in the study, which was directed by Stanford University assistant professor of emergency medicine Brian Suffoletto, to record participants’ voices both before and after drinking. Voice patterns were then analyzed to predict participants’ levels of drunkenness.

Surprising Accuracy

In this study, eighteen participants (aged 21 and older) were given an alcohol dosage depending on their weight and instructed to perform tongue twisters. Prior to and for up to seven hours after drinking, smartphone recordings were made every hour.

The researchers examined elements like pitch and frequency using computer algorithms. Remarkably, when compared to breath alcohol readings, the model built using speech patterns correctly predicted drunkenness with 98% of the samples.

Potential for Preventing Injuries

Prof. Suffoletto, the primary researcher, was taken aback by how accurate the results were and believes they may be used as a preventative measure against alcohol-related harm and fatalities. In his ideal world, cellphones would passively gather data, monitoring shifts in people’s daily routines to spot possible drinking episodes and offer prompt treatments.

Application of Advanced Technologies

Prof. Suffoletto credits the use of state-of-the-art developments in machine learning, acoustic analysis, and signal processing for the higher accuracy of their model. He thinks that this technology may play a major role in reducing the number of alcohol-related mishaps and injuries.

Just-in-Time Interventions

The project aims to provide “just-in-time interventions,” which use cellphones to notify people when they might want assistance. According to Prof. Suffoletto, integrating speech pattern sensors with other actions like walking and texting may yield a more accurate assessment of intoxication levels.

Future Studies and Recommendations

Voice pattern analysis might have a significant influence on smartphones and smart speakers, the researchers note, but larger investigations are need to validate the validity of their findings. Professor Suffoletto emphasizes that the ultimate objective of creating an intervention system to prevent injuries and save lives is to push the National Institutes of Health to create data archives for these digital indicators.


The study represents a possible advance in the use of smartphone technology to identify alcohol intoxication. The unexpected precision attained in gauging the degree of drunkenness through alterations in vocal patterns provides a platform for creative interventions that may have a substantial effect on public safety. As technology develops, including these functions into commonly used gadgets might be essential for encouraging safe drinking and lowering the number of alcohol-related accidents.

Follow our Twitter Account for Daily Insights on Technology

Leave a Comment

Your email address will not be published. Required fields are marked *