[Retracted] Application of Lightweight Deep Learning Model in Vocal Music Education in Higher Institutions

Table 2

Network structure parameter.

Type	Convolution kernel size/Step size	Number of convolution kernels	Output
Input	—	—	4484483
Conv1	33/2	128	224224128
Pool1	22/2	—	112112128
Conv2	33/1	256	112112256
Pool2	33/1	—	5656256
Conv3	33/1	256	5656256
Pool3	33/2	—	2828256
Conv4	11/1	128	2828128
Conv5	33/1	128	2828128
Pool4	33/2	—	1414128
Inception1	—	64/64, 96/16, 32/32	1414224
Pool5	33/2	—	77224
Inception2	—	64/64, 96/16, 32/32	77224
Pool6	33/2	—	44224
Inception3	—	64/64, 96/16, 32/32	44224
Pool7	22/1	—	22224
Inception4	—	64/64, 96/16, 32/32	22224
Pool8	22/1	—	11224
Output	11/1	—	1164