Intelligent fault diagnosis has recently become a prominent research area for detecting faults in rotating machinery. In this study, an end-to-end scheme utilizing a one-dimensional convolutional neural network (1D-CNN) is developed to accurately classify the faults in rotating machinery. The developed 1D-CNN model used wide kernels in the initial convolutional layer to overwhelm the impact of noise at higher frequencies and small kernels in the subsequent layers to achieve superior feature representation. Moreover, the batch normalization algorithm is incorporated after the convolutional layer to expedite the training process, and employ dropout regularization techniques after the fully connected layer to prevent overfitting. The softmax classifier is employed for fault classification, utilizing categorical cross-entropy as the loss function to minimize classification errors. The effectiveness of the proposed 1D-CNN model is evaluated using benchmark roller-bearing vibration datasets. The developed 1D-CNN model is also compared with the CNN and deep convolutional neural networks with wide first-layer kernels (WDCNN) models. The obtained results confirm that the developed method achieves 100% fault diagnosis accuracy with a simple network architecture, highlighting its superiority.