Gearbox is an important component in machine system. Hence, it is important to predict and maintain the performance of the gear system, since any unpredictable failure in this system can seriously threaten human lives and cause significant economic losses. Therefore, it is essential to inspect the gear teeth at regular intervals for identifying the crack propagation or other damages affecting the system in advance at the early stage. In this study, a new method have been proposed by the researchers which utilized artificial intelligence processes for routine maintenance. A new 3rd generation Artificial Neural Network (ANN) has been used for diagnosing and classifying the faults occurring in the spur gear systems based on the Acoustic Emission (AE) signals. Hence, they developed a test rig with various seeded gear faults, which later processed the acquired AE signals by utilizing the pre-processing technique based on the Slantlet Transform (SLT), feature extraction and Information Gain (IG) processes. These processes were used before the use of a feature selection technique which was used for developing the Spiking Neural Network (SNN) diagnosis and classification model. These processes were run using a Graphical User Interface (GUI) for fault diagnosis and classification of spur gears. Results of the study showed that this process could improve the accuracy of the diagnosis system depending on the features and information which was fed to the model. In this study, the researchers investigated the probability of increasing the accuracy better for the spur gear fault diagnosis with the help of the Spiking Neural Network (SNN) process. They achieved an accuracy of �95 when using SNN. Finally, it was concluded that the proposed technique was as an effective tool for diagnosing and classifying the faults identified in the spur gears. © 2022 IEEE.