AI taste buds are better at identifying what’s in food than you are
Selecting individual ingredients from a dish can be a fun but difficult part of a meal. Professional chefs and food scientists can spend years refining their palette. Now a robot might be able to join in on the activity thanks to the researchers behind a robot taster that combines AI with an electronic tongue that can detect small taste differences.
The Penn State research team has published a paper detailing how the AI ββ”brain” uses the artificial tongue to detect how much water is in a cup of milk, the mix of beans in a coffee blend, and even the onset of rot in fruit juice. are impossible for a human to discover.
Electronics to identify components in a mixture is not a new idea. This way, machines can measure things like acidity and temperature. But what the researchers have done goes further by using AI to mimic the way your tongue, nose and brain interpret the taste of things, beyond simple detection of pH balance. Using the advanced sensors known as ISFET (graphene-based ion-sensitive field-effect transistor), the electronic tongue can measure many complex chemicals simultaneously instead of requiring multiple types of sensors such as a thermometer and a pH test stick.
The sensors produce a huge amount of data, which can take standard computer processors some time to sort through, and the analysis wouldn’t tell you much about how watered down the milk is or how freshly squeezed your orange juice is. Instead, the researchers used AI in the form of a neural network that can mimic some of the way humans process taste.
Taste AI
After the AI ββlearned how different chemicals affect the electronic tongue’s sensors, the neural network was able to accurately identify different types of soda and the freshness of juice more than 80% of the time. However, that was just the beginning. When the scientists metaphorically let the AI ββloose and figure out its own way to analyze the data, the AI’s accuracy skyrocketed to 95%, almost never getting a wrong answer.
The combination of measuring subtle aspects of food and using AI to judge what they mean is an impressive simulation of how people taste things. This can also happen if a difference is too subtle for human perception, for example if milk is not yet bad, but soon will be.
Food testing for purity and freshness are just some of what an accurate AI tongue could do for humans. Taste is, at its most basic level, a way to identify chemicals. This means that the AI ββtaster cannot only help in the kitchen. In theory, it could help in industrial factories or medical diagnostics, detecting biomarkers of disease or changes in your health. These concepts are still in the early stages of discussion, but the electronic tongue could be a taste of the future.