Skip to main content
EURASIP Journal on Audio, Speech, and Music Processing

Table 7 Comparison of the event detection results obtained by the best single-task networks and the best double-task network

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

  

Accuracy (%)

False positive (%)

False negative (%)

Speech

Single-task network

83.99

10.41

5.60

 

Double-task network

83.81

10.67

5.52

Music

Single-task network

84.20

9.86

5.94

 

Double-task network

84.16

8.68

7.16

Back to article page