Research was conducted to design an electronic tongue based on MCU and ion-selective electrodes, software based on machine learning technology. Ion-selective electrodes were used to compose a sensors-array. MCU and A/D chip were used to compose data acquisition system. In Matlab environment, principal component analysis (PCA), back propagation neural network (BPNN), probability neural network (PNN) and support vector machine (SVM) were used to achieve pattern recognition of four kinds of drinking water and five kinds of ciders. The results showed that this electronic tongue system displayed great application potential in drinking waters and ciders analysis and also had good expansibility.