If we consider the tale of Peter Piper and his preserved peppers or a lady vending seashells by the seashore, saying challenging phrases can sound quite dissimilar after consuming alcohol.
Currently, scientists hypothesize that alterations in pitch and frequency may serve as a means to notify individuals of their degree of drunkenness.
The lead researcher of a research paper by Stanford University, Dr Brian Suffoletto, stated that the method has various potential uses in the future. One example is as an ignition lock for vehicles, where the driver would need to pass a “voice challenge” before starting the car. This technology could also be implemented in high-risk workplaces, such as school bus drivers or heavy equipment operators, to ensure the safety of the public.
A potential use of this could be in dining establishments or drinking establishments, where a bartender could receive notifications when it is appropriate to stop serving a customer.
In the Journal of Studies on Alcohol and Drugs, Suffoletto and his team describe a study in which 18 participants over the age of 21 were first observed attempting a difficult phrase.
Every individual was provided with a significant amount of alcohol based on their weight, enough to make them intoxicated. They were then instructed to say a different tongue-twister every hour for up to seven hours after consuming the alcohol. The recordings of their speech were captured.
Suffoletto explained that the tongue-twister was utilized as a means of evaluating vocal stress and uncovering any changes that may not be apparent when speaking normally.
The researchers collected breath alcohol levels from the participants before and at 30-minute intervals following the consumption of the drink, up to seven hours after.
The rhymes were recorded in one-second sections and the vocal characteristics of pitch and frequency were examined. A subset of this data, along with corresponding breath alcohol levels, was used to train an AI system. The remaining voice data was then used to test the system.
The findings indicate that the voice analysis accurately predicted alcohol intoxication with a 98% success rate. Intoxication was determined by a breath alcohol concentration of 0.08% or higher, which is the legal limit for driving in the US.
However, the study has restrictions. It solely consisted of white participants and did not examine other aspects of speech that could potentially be impacted by alcohol, such as volume.
Suffoletto noted that effective harm prevention messaging involves strategic timing. For example, reminding someone of their drinking limits as they begin to drink can have a strong impact. However, as they become heavily intoxicated, the effectiveness of these interventions decreases.
Petra Meier, a public health professor at the University of Glasgow who did not participate in the study, emphasized that the research was conducted on a small scale and under strict conditions.
“I am optimistic about the possibility of promising advancements that could potentially have practical implications. However, it would be necessary to initially conduct experiments on a larger and more varied group before implementing this approach,” she stated, emphasizing the importance of testing the approach in real-world scenarios.
Meier explained that the test should be conducted on both light and heavy drinkers to ensure its efficacy. This is because heavy drinkers may not exhibit changes in their speech until they are highly intoxicated, leading them to mistakenly believe they are sober enough to drive.