Table 8 Model parameters for different audio pattern recognition tasks.

From: Dense dynamic convolutional network for Bel canto vocal technique assessment

Dataset

Number

Dimension

Stride

Kernel Size

Group

MVSet11

4

32–64–128–256

1–2-2-2

1–3-3-3

1–3-3-1

Urbansound8 K41

4

32–96–160–256

1–2-2-2

1–3-3-3

1–3-3-1

FSD50 K8

4

32–96–160–256

1–2-2-2

1–3-3-3

1–3-3-1

GTZAN2

4

32–64–128–256

1–2-2-2

1–3-3-3

1–3-3-1

Back to article page