There is compelling evidence from animal experiments that noise exposure suppresses testosterone in males by affecting the hypothalamic-pituitary-testicular axis. Virtually nothing is known about its effect in humans. Therefore, the aim of this study was to explore the association between occupational noise exposure and serum testosterone in a representative sample of the general population. The sample has been taken from the National Health and Nutrition Examination Survey (NHANES) data for the period between 1999 and 2004 and is limited to employed men aged 16-85+ years at the time. The associations between noise exposure (either established according to the Occupational Information Network -O*NET categories or self-reported) and total and free testosterone (TT and FT, respectively) were analysed using linear regression models with increasing adjustments. In the fully adjusted model (n=414), the third quartile of the O*NET noise exposure was associated with lower TT and FT, which reached statistically significant decrease of -58.32 ng dL -1 (95 % CI: -111.22, -5.42) and -1.58 ng dL -1 (95 % CI: -2.98, -0.18), respectively. In stratified analyses, younger, lower income, normal weight, better hearing, and workers not using hearing protection at work experienced significantly more severe adverse effects than the rest. The odds for hypogonadism (TT<300 ng dL -1 ) did not significantly rise with one interquartile range increment in O*NET noise exposure (OR=1.24, 95 % CI: 0.64, 2.39). Self-reported loud noise exposure did not significantly decrease TT when all men were considered (n=214) and only in the men ≥37 years did it decrease TT significantly by -87.55 ng dL -1 (95 % CI: -158.35, -16.74). In conclusion, noise exposure was associated with lower TT and FT only in some population subgroups and these associations were non-linear.