The new app that could tell if you are too drunk to drive home

A new app could tell you if you are too drunk to drive by listening to your voice as you recite tongue twisters.

A phone’s microphone can be used to determine a user’s level of intoxication and the app made by Stanford University scientists is 98 per cent accurate, according to a new study.

A small study of 18 adults involved getting people drunk on a specific amount of alcohol and then asking them to recite random tongue-twisters before and after drinking. Tongue-twisters used in the study included Peter Piper, She sells sea shells, and Woodchuck.

A person’s voice was recorded by a phone placed on a table about two feet away and an app analysed their voice patterns to predict intoxication levels.

Changes to voice patterns are a known consequence of alcohol intake, but reliable methods to measure and track it have yet to be developed.

Dr Brian Suffoletto, an associate professor of emergency medicine at Stanford and the lead researcher, said: “The accuracy of our model genuinely took me by surprise.

“While we aren’t pioneers in highlighting the changes in speech characteristics during alcohol intoxication, I firmly believe our superior accuracy stems from our application of cutting-edge advancements in signal processing, acoustic analysis and machine learning.”

The tool could be used to monitor levels of drunkenness over an evening, the scientists say, or as a “just-in-time” measure that stops a person driving a car or motorbike if they are too drunk when they turn the engine on.

Deployment of the technology would allow people to be responsible, but will have to be careful to not be too burdensome or irritating and become an inconvenience.

“While one solution could be to frequently check in with someone to gauge their alcohol consumption, doing so could backfire by being annoying (at best) or by prompting drinking (at worst),” Dr Suffoletto added.

“Imagine if we had a tool capable of passively sampling data from an individual as they went about their daily routines and surveil for changes that could indicate a drinking episode to know when they need help.”

Other easy to use tests, such as using a phone to monitor walking and texting competence, could be combine with the voice analysis as well in the future to boost accuracy and reliability.

“An individual might not speak for hours, but they could be walking. There might be instances where they’re stationary at a bar, neither walking nor talking, yet actively texting,” Dr Suffoletto said.

The product is not yet commercially available and will undergo trials on a larger number of people, with more diversity in accents, before being made public.

Researchers are now planning to build on the results of the study and hope to “develop and test technologies that leverage digital biomarkers” to predict a person’s level of drunkenness.

The study is published in the Journal of Studies on Alcohol and Drugs.