The effects of high hydrostatic pressure on volatile generation in milk were investigated in this study. Raw milk samples were treated under different pressures (482, 586, and 620 MPa), temperatures (25 and 60 degrees C), and holding times (1, 3, and 5 min). Samples submitted to heat treatments alone (25, 60, and 80 degrees C for 1, 3, and 5 min) were used for comparison. Trace volatile sulfur compounds were analyzed using solid-phase microextraction (SPME) and gas chromatography (GC) with pulsed-flame photometric detection (PFPD), whereas the rest of the volatile compounds were analyzed using SPME-GC with flame ionization detection (FID). Multivariate analysis of variance (MANOVA) and principal component analysis (PCA) were used to study the effect of pressure, temperature, and time on volatile generation. Relative concentration increases of 27 selected volatile compounds were compared to an untreated sample. It was found that pressure, temperature, and time, as well as their interactions, all had significant effects (P < 0.001) on volatile generation in milk. Pressure and time effects were significant at 60 degrees C, whereas their effects were almost negligible at 25 degrees C. The PCA plot indicated that the volatile generation of pressure-heated samples at 60 degrees C was different from that of heated-alone samples. Heat treatment tended to promote the formation of methanethiol, hydrogen sulfide, methyl ketones, and aldehydes, whereas high-pressure treatment favored the formation of hydrogen sulfide and aldehydes.