Abstract: To build environment-friendly high-altitude UHV AC transmission lines, it is of great significance to study the long-term audible noise (AN) observation data of the completed transmission lines. The AN in bad weather is mixed with wind, rain, snowfall, and other factors, and it is difficult to reveal the influence law of AN. Therefore, this paper uses the test results of the Guanting electromagnetic environment observation station for seven months to analyze the environmental influence law of AN and research its prediction model in fair weather. The experimental results show that: (1) AN has a strong positive correlation with ambient temperature and wind speed in fair weather. In particular, when the wind speed is lower than 2.2 m/s, the wind speed has a negligible effect on the audible noise; when the wind speed is greater than 2.2 m/s, the audible noise has an increasing trend with the increase of the wind speed. (2) Due to condensation and frost, the rise of AN in the morning is ahead of the increase in temperature and wind speed, showing the phenomenon that AN is not synchronized with the change of environmental factors. (3) The proposed CNN-LSTM prediction model's RMSE in three seasons is 1.934 3, 1.221 5, 1.248 8, and the MAE is 1.425 0, 0.910 0, 0.942 5, respectively, which achieves a more accurate prediction.