Abstract: Sound sensors are easy to install, do not need to adhere to the sensor and do not interfere with the machining process, so it has a wide application prospect in the online monitoring of tool wear in the machining process. However, the sound propagation process in the closed space of the machine tool is affected by the inner wall of the machine tool, spindle, table, and other structural effects, which will produce reflection, diffraction, and other phenomena, resulting in greater variability in the distribution of the acoustic field, which will bring trouble to the reasonable arrangement of the sound sensor. Therefore, the numerical control milling machine is taken as the research object, the sound field simulation model is established based on the acoustic finite element theory, the pressure acoustic transient analysis method is used to establish the sine acceleration sound source signal propagation model in the machine tool, and the Fourier transform is used to analyze the sound field in the machine tool spectrum. The results show that the sound pressure is larger at the position 0.4 m away from the sound source and the wall surface, and decreases with the increase of the sound source distance in the middle of the machine tool. This model lays the foundation for tool wear monitoring based on acoustic signals.