Published Paper Details:

This study attempts to give the user the machine's operational condition while simultaneously detecting flaws or damage in big industrial machinery in real-time. However, due to the complexity of real-world systems and the clear presence of nonlinear elements, it is highly challenging to diagnose a machine malfunction using standard approaches based on mathematical models. A computer-aided diagnostic system with artificial intelligence (CADS-AI), which aids in early fault diagnosis, can resolve this issue. To assure that the CADS solution can be trusted, a machine failure detection system based on sound patterns is created in this study. The YAMNet Model classified the sounds as normal or abnormal with an accuracy of 78.31%. By using window inference on the audio file with a mean score, the predicted output was decoded. By using real-time audio input or uploading machine sounds to the system, the proposed solution can be inferred.