Audio Detection using Mel-frequency Cepstral Coefficients

The lack of benchmark findings for comparison with any suggested approach is one of the most fundamental challenges in sound event detection research. Distinct research explore different sets of events and datasets, making it difficult to distinguish between new and existing methods. We calculated the average accuracy and mean AUC using the ESC 50 dataset in this research. We employ a Gaussian Mixture model-based component depiction explicitly and solidify it with linear and non-linear Support Vector Machines.