Rainfall prediction is a challenging task due to its high variability and uncertainty. Artificial intelligence techniques like wavelet neural networks (WNN) have been used to predict precise rainfall timing using wavelet activation functions. This study focuses on predicting rainfall using WNN based on meteorological inputs. The experimental findings show that both DWT-based ANN and Haar wavelet-based WNN models are excellent at predicting rainfall, representing connections and temporal dependencies between meteorological signals and rainfall patterns. The achieved RMSE values show accurate forecasts, suggesting that these models could be useful tools for tasks involving rainfall prediction. The combined use of artificial neural networks and wavelet analysis presents a possible approach for rainfall prediction based on meteorological signals, allowing for flexibility in capturing temporal dependencies and increasing precision. These models could help better understand and predict rainfall patterns, potentially impacting various applications in agriculture and water resource management.