In this work, an electronic tongue (ET) system is developed for the discrimination of infusions and teas made from different herb leaves. The ET is based on an array of four non‐specific potentiometric sensors plus principal component analysis (PCA) data processing. The four sensors used are a cation‐selective electrode and an anion‐selective electrode, containing both a plasticized polymeric membrane with different ionic additives, a commercial pH electrode and a redox electrode. These provide simple, rapid and very reproducible potentiometric responses after direct exposure to infusions and teas, which do not need any pre‐treatment. A radar plot of potentiometric values for each sensor, obtained after sixty seconds of array exposure to the samples, revealed a differentiated response for each sensor and infusion. This was especially marked for anion‐selective and redox electrodes. The PCA results of the potentiometric signals point to certain trends that can be used for classification purposes. Although a small number of sensors is used in the developed ET, interesting classification functions have been achieved, which will be exploited in the near future with a larger number of samples. The concept of an ET formed by a reduced number of potentiometric sensors is starting to spread to other types of beverages.