With the huge growth and complexity of seismic data, manual labeling of seismic facies has become a significant challenge. Recently, deep learning algorithms (particularly CNNs) have been used to simplify this task. In this submission, we demonstrate the advantage of using advanced signal processing techniques to preprocess signals before feeding them to deep learning algorithms. Our approach combines maximal overall discrete wavelet transform with recurrent neural networks (RNN) to improve the automated seismic facies analysis. This proposed framework generates more accurate results in a more efficient way. The combination of RNN with wavelets achieves more accurate results than just using CNNs. The results were demonstrated in a recent hackathon organized by SEAM AI where the MathWorks® team was ranked at the top – our F1 scores on test data were significantly higher than other teams