Skip to main content

Table 12 Top-10 negative distractor events for music (event labels related to false negative decisions of the network about the “Music” class)

From: Exploring convolutional, recurrent, and hybrid deep neural networks for speech and music detection in a large audio dataset

Event

Event ID

Ratio

\(d^{-}_{mu}\)

Animal

/m/0jbk

32/84

0.230

Speech

/m/09x0r

1266/5617

0.223

Inside, small room

/t/dd00125

49/166

0.221

Vehicle

/m/07yv9

39/135

0.205

Domestic animals, pets

/m/068hy

19/40

0.199

Dog

/m/0bt9lr

16/27

0.194

Outside, rural

/t/dd00129

22/61

0.190

Flute

/m/0l14j_

13/17

0.180

Outside, urban

/t/dd00128

20/63

0.169

Television

/m/07c52

15/40

0.157

  1. \(d^{-}_{mu}\) score (Eq. 14) is used to rank the events. The ratio column shows the number of false negatives for music where the distractor event label is found (numerator) and the number of music segments that contain the distractor event (denominator)