Abstract A home‐made hybrid electronic tongue was set up, validated and applied to discriminate soft drinks fortified with plant extracts of green tea. The e‐tongue consists of a flow injection system equipped with two electrochemical and one optical sensors. Different formulations of soft drinks composed of glucose and epigallocatechin gallate were then discriminated by principal component analysis. Furthermore, two partial least squares regression models were developed to estimate the “sweetness” ( r 2 of 0.992) and “bitterness” ( r 2 of 0.993) of the model solutions and commercial soft drinks, before and after their fortification with epigallocatechin gallate.