Kernel Shape in a CNN Audio Model
(Code on GitHub.) Audio has a strong temporal component. Unlike an image, audio is a thing that happens in time, not an arrangement of items in a space. And yet many audio classification models treat spectrograms as if they were still images and not events, an artifact of early successes applying visual models to audio datasets. I took the ESC-50 dataset, created a simple five-layer CNN model, and trained it with various kernel shapes and sizes. My hypothesis: kernels that extend more in the temporal dimension will have better performance. ...