AI Outperforms Human Experts in Distinguishing Whisky Origins and Aromas

AI has just achieved another milestone: it can now distinguish whiskies better than you - Softonic

Are the discussions about which whisky we are drinking about to end? Clearly not Another victory for artificial intelligence. An AI-powered algorithm has identified the five strongest notes of each analyzed beverage with more precision than any member of a panel of expert tasters. The researchers have used this technology to predict the notes that whisky releases and determine whether a drink has been made in the United States or Scotland, two of the places where the most whisky is produced in the world, with differences ranging from the type of production to the ingredients used. The work represents a step forward towards automated systems capable of predicting the complex aroma of whisky from its molecular composition. Expert panels often evaluate woody, smoky, buttery, or caramelized aromas, which can help ensure they do not vary substantially between batches of the same product. Determining the aroma of a whisky is not an easy task. Most of the stronger notes of the spirit are a complex mixture of chemical substances that interact in the nose and even mask each other to create a particular aromatic impression. These interactions make it extremely difficult to predict how the whisky will smell based on its chemical signature. For the latest work, the researchers obtained the chemical composition of 16 American and Scottish whiskies, including Jack Daniel's, Maker's Mark, Laphroaig, and Talisker, and details of their aromas from a panel of 11 experts. The information was used to train artificial intelligence algorithms to predict the five main aromas and the origin of the beverages from their molecular components. One of the algorithms had an accuracy of over 90% in distinguishing American alcoholic beverages from Scottish ones, although the performance would likely decrease with beverages it had not been trained on. On average, it identified the five strongest notes of each whisky with greater accuracy and consistency than any of the expert panel members. The results have been published in Communications Chemistry. The compounds menthol and citronellol helped identify American whiskies, which often have a caramel note. Methyl decanoate and heptanoic acid were important for identifying Scotch whisky, which usually has a smoky or medicinal aroma. Researchers see applications that go beyond whisky, from detecting counterfeit products through discrepancies in their smell to finding the best way to mix old recycled plastics, which can develop unpleasant odors, into new products without the odor being noticeable.

AI learns to distinguish between aromas of US and Scottish whiskies

One algorithm identified the five strongest notes in each drink more accurately than any one of a panel of experts Notch up another win for artificial intelligence. Researchers have used the technology to predict the major notes that waft off whisky and determine whether a dram was made in the US or Scotland. The work marks a step towards automated systems that can predict the complex aroma of whisky from its molecular makeup alone. Expert panels usually assess woody, smoky, buttery or caramel odoursaromas, which can help to ensure aromas don't vary substantially between batches of the same product. "The beautiful thing about the AI is that it is very consistent," said Dr Andreas Grasskamp, who led the research at the Fraunhofer Institute for Process Engineering and Packaging in Freising, Germany. "You have this subjectivity still in trained experts. We are not replacing the human nose with this, but we are really supporting it through efficiency and consistency." Nailing down a whisky's aroma is no simple business. Most of the strongest notes in the spirit are a complex mixture of chemicals that interact in the nose and even mask one another to create a particular aromatic impression. The interactions make it extremely difficult to predict a whisky's rich aroma from its chemical signature. For the latest work, the researchers obtained the chemical makeup of 16 US and Scottish whiskies, including Jack Daniels, Maker's Mark, Laphroaig and Talisker, and details of their aromas from an 11-strong expert panel. The information was used to train AI algorithms to predict the five major aromas and origin of the whiskies from their molecular constituents. One algorithm was more than 90% accurate at distinguishing the US from Scottish whiskies, though the performance would be likely to drop against tipples it had not been trained on. On average, it identified the five strongest notes in each whisky more accurately and consistently than any individual on the expert panel. Details are published in Communications Chemistry. The compounds menthol and citronellol helped to identify US whiskies, which often have a caramel-like note. Methyl decanoate and were important for identifying scotch, which often has a smoky or medicinal aroma. The researchers see applications in areas beyond whisky, from detecting counterfeit products through discrepancies in their smell, to finding the best ways to blend old recycled plastics, which can develop unpleasant odours, into new products without the smell being noticeable. Dr William Peveler, a senior lecturer in chemistry at the University of Glasgow, said the approach could provide more "stability" than a human taste panel. "The flavour notes of a whisky brand could be quickly checked from batch to batch or blend to blend based on the chemical signature alone to try and ensure a consistent house style," he said. The study involved only a small number of whiskies and it is unclear how the AI would perform when faced with more, he added, and because the flavour notes developed with age in the cask. "The other thing with whisky is that perception of flavour is hugely influenced by the environment it's consumed in and other external factors, so there could be some work to do on other factors that influence flavour perception and prediction in such an emotive product," he said.

AI beats human experts at distinguishing American whiskey from Scotch

Using descriptions of flavours or chemical data, artificial intelligence can tell apart whiskies from different countries and identify their constituent aromas Artificial intelligence can tell Scotch whisky from American whiskey and identify its strongest constituent aromas more reliably than human experts - by using data rather than tasting the drinks. Andreas Grasskamp at the Fraunhofer Institute for Process Engineering and Packaging IVV in Germany and his colleagues trained an AI molecular odour prediction algorithm called OWSum on descriptions of different whiskies. Then, in a study involving 16 samples - nine types of Scotch whisky and seven types of American bourbon or whiskey - they tasked OWSum with telling drinks from the two nations apart based on keyword descriptions of their flavours, such as flowery, fruity, woody or smoky. Using these alone, the AI could tell which country a drink came from with almost 94 per cent accuracy. Because the complex aroma of these spirits is determined by the absence or presence of many chemical compounds, the researchers also fed the AI a reference dataset of 390 molecules commonly found in whiskies. When they gave the AI data from gas chromatography-mass spectrometry showing which molecules were present in the sample spirits, it boosted OWSum's ability to differentiate American from Scotch drams to 100 per cent. Compounds such as menthol and citronellol were a dead giveaway for American whiskey, while the presence of methyl decanoate and heptanoic acid pointed to Scotch. The researchers also tested both OWSum and a neural network on their ability to predict the top five odour keywords based on the chemical contents of a whisky. On a score from 1 for perfect accuracy to 0 for consistent inaccuracy, OWSum achieved 0.72. The neural network achieved 0.78 and human whisky expert test participants achieved only 0.57. "[The results] underline the fact that it's a complicated task for humans, but it's also a complicated task for machines - but machines are more consistent than humans," says team member Satnam Singh, also at the Fraunhofer Institute. "But that's not to say that humans are not needed: we do need them to train our machines, at least, right now." Neither model takes into account the concentration of molecules, only their absence or presence, which is something the researchers hope to rectify, and which may yield even greater accuracy. Grasskamp says such AI tools could be used for quality control in distilleries, or to help develop new whiskies, as well as detecting fraudulent ones. But they could also be used for "anything that smells", such as other food and drink production or in the chemical industry.

AI Predicts Whisky Aromas and Origins with Over 90 Percent Accuracy

Consistent and efficient analysis surpasses expert panels Efforts to use artificial intelligence in analysing whisky aromas have yielded promising results, as researchers demonstrated the technology's ability to identify key notes and origins of whiskies. The study, conducted by the Fraunhofer Institute for Process Engineering and Packaging in Germany, explored the molecular makeup of 16 US and Scottish whiskies, including brands such as Jack Daniel's, Maker's Mark, Laphroaig, and Talisker. According to reports, the findings indicated that AI systems could provide consistency and precision in determining whisky aromas, surpassing human expert panels in certain aspects. The research, published in Communications Chemistry, involved training algorithms using chemical compositions and aroma profiles provided by an 11-member expert panel. The AI was tasked with predicting the five most prominent aroma notes and distinguishing between US and Scottish whiskies. It reportedly achieved over 90 percent accuracy in identifying the origins of the whiskies, although this figure is expected to decrease when applied to untrained samples. Dr Andreas Grasskamp, the study's lead researcher, highlighted the AI's consistent performance, stating to The Guardian that it serves to complement, rather than replace, human assessments. The analysis pinpointed compounds such as menthol and citronellol in US whiskies, known for their caramel-like notes, and methyl decanoate and heptanoic acid in Scottish whiskies, associated with smoky and medicinal aromas. The research is expected to have broader applications beyond whisky analysis. Reports have suggested that the technology could aid in detecting counterfeit products and managing odours in recycled plastics. Dr William Peveler, a senior lecturer at the University of Glasgow, noted to The Guardian that such approaches could offer stability in maintaining consistent flavour profiles across whisky batches. While the study demonstrated potential, limitations remain, such as the small sample size and the challenges posed by flavour changes during aging. Experts believe further work is necessary to account for environmental and sensory factors influencing whisky perception.

AI can tell Scottish and American whiskies apart

Artificial intelligence (AI) can be used to tell the difference between American whiskey and Scotch, and identify their prominent aromas, potentially better than a human expert. More than 40 different compounds with diverse structures determine the aroma profile of whisk(e)y. Chemical interactions of these compounds in the olfactory system make it very challenging to assess or predict the characteristics of a whisk(e)y's aroma based solely on its molecular composition. Human panels are widely used to evaluate flavours, however, because olfactory perception can be subjective, such evaluation methods require a significant amount of time, money and often, trained panellists. Researchers in Germany therefore set out to see if machine learning methods could be used to assess the molecular composition of seven American and nine Scotch whiskies. To do this they used two machine learning algorithms; OWSum, a molecular odour predictions algorithm developed by the researchers, and a convolutional neural network. The team analysed the whisk(e)y samples with GC-MS and automatic compound detection analysis. They also collected sensory data, generated by a human expert panel using an evaluation method called 'rate-all-that-apply' (RATA), to determine the top five odour descriptors per sample. 'Caramel-like' was the most characteristic odour descriptor for American whiskey, whereas 'apple-like', 'phenolic' and 'solvent-like' odours were more pronounced in Scotch samples. This, the researchers said, 'show that there are clear relationships between volatile molecules as well as olfaction with the type of whisky'. They then used the algorithms to find out whether they could predict the smell of a whisk(e)y, represented by the top five odour descriptors and found that they were able to do so with 'promising accuracy'; OWSum was able to determine whether a sample was American whiskey or Scotch with an accuracy of over 90% and both algorithms were able to identify the five strongest notes of a specific whisk(e)y more accurately and consistently on average than any individual human expert. 'OWSum not only offers a method to quickly classify whiskies, but also allows us to analyse their ingredients or characteristic features in one step,' they said.