Tryptophan is one of the essential amino acids in human, and also considered as an important limiting amino acid in rice. Alkali hydrolysis method was used to determine the tryptophan content of 272 representative samples selected from the 1 256 samples in four years. Based on the milled rice flour spectroscopy, by comparing the prediction abilities of the equations with global calibrations and local calibration, the global calibration method with MPLS and the local calibration method were better than others. Their calibration models could be introduced to application because of low standard errors of prediction for both methods (0.007%) and high coefficients of determination (87.1% and 87.4%, respectively). Furthermore, the prediction ability of the equation based on the brown rice spectroscopy could be still used for determination application though their effects were worse than that of the milled rice flour spectroscopy. It was concluded that the equation could be directly used to estimate the content of tryptophan in milled rice, and the technology of near infrared reflectance spectroscopy developed in this study could be applied in selecting breeding materials for higher tryptophan content and in quality control during rice processing.