In order to study the quality change of fermented millet pepper, the sensory evaluation model of fermented millet pepper was established using the intelligent sensory instruments (electronic nose and electronic tongue) and sensory evaluation, combinating with the method of correlation analysis, principal component analysis, stepwise regression analysis and fitting test. The results showed that the weight of the simplified sensory indicators were 38.96%, 35.06% and 25.97%, respectively, and the fitting test R²=0.998 indicated that the model can replace the original evaluation method. Taking the PCA of the intelligent sensory index as the independent variable and the sensory evaluation model as the dependent variable, the principal component regression equation was established by factor analysis: Y (sensory score) =67.672＋8.296 × color PC1－1.571 × texture PC1－3.431× PC1 electronic nose＋0.288×PC1 electronic tongue, and the variance inflation factor test showed that the model had no obvious multicollinearity (VIF＜10). The new quality evaluation model can fit the change of sensory quality during fermentation process of millet pepper, and offset the defect of the reproducibility and stability of the artificial sensory.