Artificial Taste System: The Era of Digital Taste with Graphene Oxide Technology

Metaverse Planet September 1, 2025Last Updated: September 1, 2025

0 2 minutes read

Scientists in Beijing have developed a graphene oxide-based artificial taste system that mimics the sense of taste. The technology can distinguish between tastes with 99% accuracy. Here are the details:

Researchers from the National Center for Nanoscience and Technology in Beijing have developed an artificial taste system that is set to revolutionize the digitalization of the sense of taste. This innovative graphene oxide-based technology mimics the taste perception mechanism of the human brain, detecting tastes such as sour, salty, bitter, and sweet with almost 99% accuracy.

The researchers were inspired by neuromorphic engineering in developing their system. This approach models how biological taste buds and neurons work together, allowing the device not only to sense tastes but also to learn them. In this way, the system can store and recall tastes from its memory, just like the human brain.

Is it Possible to Digitize Tastes?

The sense of taste is known to be one of the most difficult senses to digitize compared to other senses like sight and hearing. The main reason for this is that taste perception occurs through ion movements. To overcome this challenge, the Beijing team developed a special structure called the Graphene Oxide Ionic Sensory Memristive Device (GO-ISMD).

Within the device’s nano-scale channels, ions undergo adsorption and desorption processes, generating electrical responses. This mechanism allows the system to behave like a biological taste bud. In other words, the device can both sense tastes and process and store them in its memory. This feature is one of the most important factors that distinguishes the technology from traditional sensors.

The research team used the reservoir computing method to optimize this process. In this approach, the device converts the electrical signals it receives into unique digital patterns. These patterns are then transferred to a specially trained single-layer neural network. As a result, the device can learn, recall, and classify different tastes and aromas with high accuracy.

The system was tested in a laboratory setting on four basic tastes: sour (acetic acid), salty (NaCl), bitter (MgSO₄), and sweet (lead acetate). The data transferred to the neural network achieved an impressively high accuracy rate of 98.5% in distinguishing between the tastes. The researchers also tested more complex flavors, such as coffee, cola, and mixtures of these beverages. The artificial taste system successfully classified these beverages as well, demonstrating significant potential for future applications in food analysis, quality control, and the beverage industry.