Skip to main content

Table 1 Comparison between the performance of different CNN architectures on the pathological voice data set in the form mean AUC ±95% confidence interval

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

Detectors CNN32 CNN64 CNN128 CNN256 CNN512
Noise 1.00 ±0.00 1.00 ±0.00 1.00 ±0.00 1.00 ±0.00 1.00 ±0.00
Distortion 0.98 ±0.01 0.98 ±0.01 0.99 ±0.00 0.98 ±0.01 0.98 ±0.01
Reverberation 0.90 ±0.01 0.91 ±0.01 0.93 ±0.01 0.91 ±0.01 0.89 ±0.01
  1. The bold-faced numbers represent the best performance