Multiple Musical Instrument Signal Recognition Based on Convolutional Neural Network

Table 1

Change process of the feature size in the benchmark model.

Input size	Operation	Output size
176 × 165 × 1	2 × 1 Convolution kernel, 352 channels	352 × 164 × 1
352 × 164 × 1	3 × 1 Maximum pooling	352 × 54 × 1
352 × 54 × 1	3 × 1 Convolution kernel, 704 channels	704 × 52 × 1
704 × 52 × 1	3 × 1 Channels	704 × 17 × 1
704 × 17 × 1	2 × 1 Channels, 704 channels	704 × 8 × 1
704 × 8 × 1	1 × 1 Channels, 11 channels	11 × 8