The prediction of environmental sound pressure levels related to the operation of wind turbines is still a concerning issue. During the last decade some different proposals have been made to enhance the results from the old prediction methods. We present the prediction model developed at the Environmental Engineering Department of the Faculty of Engineering (Universidad de la República, Uruguay), which has shown a good agreement with measured data at about ten different wind farms along the country. First of all, the atmospheric stability class according to Pasquill-Gifford and the wind velocity are taken into account for determining the acoustic power of the device. Then, a first module for describing noise generation along the blades due to turbulent phenomena (incoming wind turbulence and blades boundary layer release) is applied to compute sound pressure levels at 100 m far from the tower. Some coefficients have been experimentally obtained through wind tunnel tests. At last, a propagation module completes the prediction model to compute sound pressure levels at different distances from the tower. The propagation module considers not only geometric divergence but atmospheric absorption and turbulent dissipation (both phenomena depend on the frequency). The final sound pressure levels at a given reception point are obtained by superposing the sound pressure levels due to different wind turbines operation and the environmental noise due to the wind. Every computing is done in octave bands and the final results are expressed as LAeq values. The development of the prediction model has been done with measured data from 4 wind farms built on flat terrain. During calibration process some adjustment was needed for better prediction of noise propagation from wind farms built on complex terrain. This easy-to-use method has been programmed in Matlab®. The obtained results show almost 80 % cases within in ±3 dB range.