Noise annoyance is influenced by sound-related factors: type of noise, noise level and frequency, and person-related factors-physiological, psychological, and social factors. Prior to implementation of the Directive 2002/49/EC of the European Parliament and of the Council in Serbia, there was a need for the first comprehensive study on noise annoyance in Serbian urban population. The aim of this study was to determine principal factors for high noise annoyance in an adult urban population and to assess their predictive value. A cross-sectional study was performed on 3097 adult residents of a downtown municipality in Belgrade (1217 men and 1880 women), aged 18-96 years. Equivalent noise levels [Leq (dBA)] were measured during day, evening and night in all streets of the municipality. Noise annoyance was estimated using self-reported annoyance scale. Noise annoyance showed strong correlation with noise levels, personal characteristics and some housing conditions. Dose-response relationship was found between the percentage of highly annoyed residents and Lden. Logistic regression model identified increased risk for a high level of noise annoyance with regard to: orientation of living room/bedroom toward the street (Odds Ratio=2.60; 95% Confidence Interval=2.04-3.31), duration of stay at apartment during the day [OR=1.04, 95%CI=1.02-1.06 (per hour)], noise sensitivity [OR=1.04, 95%CI=1.03-1.04 (per scale unit)], and nighttime road-traffic noise level [OR=1.03, 95%CI=1.02-1.04 (per decibel)].