The volatile compounds of cranberries obtained from four cultivars (Early Black, Y1; Howes, Y2; Searles, Y3; and McFarlin, Y4) were analyzed by gas chromatography-olfactometry (GC-O), gas chromatography-mass spectrometry (GC-MS), and GC-flame photometric detection (FPD). The result presented that a total of thirty-three, thirty-four, thirty-four, and thirty-six odor-active compounds were identified by GC-O in the Y1, Y2, Y3, and Y4, respectively. In addition, twenty-two, twenty-two, thirty, and twenty-seven quantified compounds were demonstrated as important odorants according to odor activity values (OAVs > 1). Among these compounds, hexanal (OAV: 27-60), pentanal (OAV: 31-51), (E)-2-heptenal (OAV: 17-66), (E)-2-hexenal (OAV: 18-63), (E)-2-octenal (OAV: 10-28), (E)-2-nonenal (OAV: 8-77), ethyl 2-methylbutyrate (OAV: 10-33), β-ionone (OAV: 8-73), 2-methylbutyric acid (OAV: 18-37), and octanal (OAV: 4-24) contributed greatly to the aroma of cranberry. Partial least-squares regression (PLSR) was used to process the mean data accumulated from sensory evaluation by the panelists, odor-active aroma compounds (OAVs > 1), and samples. Sample Y3 was highly correlated with the sensory descriptors "floral" and "fruity". Sample Y4 was greatly related to the sensory descriptors "mellow" and "green and grass". Finally, an aroma reconstitution (Model A) was prepared by mixing the odor-active aroma compounds (OAVs > 1) based on their measured concentrations in the Y1 sample, indicating that the aroma profile of the reconstitution was pretty similar to that of the original sample.