Aerofoil noise not only has a negative impact on the environment but also increases energy consumption during flight. Therefore, reducing wing noise during the wing design phase is very important. This paper proposes a dynamic weight-based wing aerodynamic noise prediction model. Firstly, a wing noise prediction model is established using particle swarm optimization-based support vector regression (PSO-SVR) method. Then, genetic algorithm (GA) is used to select the best variable weights for the final noise prediction model. Finally, NASA’s NACA0012 airfoil noise data is used for verification. The results show that compared with existing methods, the proposed method has faster convergence speed and better generalization ability, and can achieve higher prediction accuracy. It can be used for noise prediction caused by low Mach number free flow and other prediction tasks lacking expert knowledge.