Skip to main content

Table 2 The performance of the CNN128 system when no parameters were shared across detectors and when the parameters in some layers were shared

From: A CNN-based approach to identification of degradations in speech signals

Detectors

Independent training

Sharing parameters

 

(first approach)

(second approach)

Noise

1.00 ±0.00

1.00 ±0.00

Distortion

0.99 ±0.00

0.98 ±0.01

Reverberation

0.93 ±0.01

0.88 ±0.01

  1. The results are in the form mean AUC ±95% confidence interval