Researchers in Germany have developed algorithms to differentiate between Scotch and American whiskey. The machines can also discern the aromas in a glass of whiskey better than human testers.
JUANA SUMMERS, HOST:
For this next story, we are talking about whiskey, so we’ve got a bottle here in the studio.
(SOUNDBITE OF CORK POPPING)
AILSA CHANG, HOST:
What? No fair. What about me?
SUMMERS: Sorry.
(SOUNDBITE OF WHISKEY POURING)
SUMMERS: I’m going to give it a whiff.
CHANG: OK. And what notes are you picking up, Juana?
SUMMERS: I am no expert, but there’s a little bit of spice, a little sweetness.
CHANG: Well, Juana, as you sniff that bouquet that I don’t get to smell, here’s what is happening in your brain, according to Andreas Grasskamp.
ANDREAS GRASSKAMP: You start with these molecules that go from the whiskey, in this case, into the air, so they can reach your nose, and you can smell them. And this leads to a nerve signal that then tells your brain, oh, it smells like that.
CHANG: Grasskamp is with the Fraunhofer Institute for Process Engineering and Packaging in Germany, and he and his colleagues have figured out another way to identify whiskey aromas – using artificial intelligence.
SUMMERS: I am intrigued. So Ailsa, what does AI think this glass of rye smells like?
CHANG: Well, it says fruity, caramel, vanilla, spicy and orange. And Grasskamp says AI is actually pretty good at evaluating whiskey.
GRASSKAMP: We were able to get on the level with the human nose, accurately saying what the sample smells like.
CHANG: They describe how this works in the journal Communications Chemistry. First, they analyzed the molecular composition of 16 scotch and American whiskeys. Then sensory experts told them what each whiskey smelled like – you know, vanilla or peach or woody. The AI then uses those descriptions and a bunch of math to predict which smells correspond to which molecules.
SUMMERS: OK. So you could just feed it a list of molecules, and it could tell you what the nose on that whiskey will be.
CHANG: Exactly. The model was able to distinguish American whiskey from scotch.
SUMMERS: Easy.
CHANG: But when it came to determining aromas, the machines actually did better than the humans, though Grasskamp emphasized that they’re not trying to replace the human nose.
SEBASTIEN GAVILLET: My name is Sebastien Gavillet, and I am a nose for hire.
SUMMERS: Gavillet is a whiskey and wine consultant with Wine Aromas Le Nez du Vin. He sees AI as a tool to augment humans’ capabilities. In fact, he already uses his own AI model, trained on decades of his own tasting notes, to analyze whiskey.
GAVILLET: AI has a trillion times the potential to analyze and break things down.
CHANG: He says AI can help develop new whiskey formulas, but he doesn’t think whiskey or wine sommeliers are going anywhere.
GAVILLET: I value my sommelier who comes to the table and propose me, you know, a type of wine to drink based on the characteristic of wines that I like. There’s communication there. There’s human touch, and there’s a certain romance in the ritual.
SUMMERS: He says the machines haven’t quite figured out the romance just yet.
(SOUNDBITE OF MUSIC)
Copyright © 2024 NPR. All rights reserved. Visit our website terms of use and permissions pages at www.npr.org for further information.
NPR transcripts are created on a rush deadline by an NPR contractor. This text may not be in its final form and may be updated or revised in the future. Accuracy and availability may vary. The authoritative record of NPR’s programming is the audio record.
