Skip to main content

Table 3 Speech denoising SAR scores

From: Blind source separation with optimal transport non-negative matrix factorization

  OT OT + OT filter KL IS E
  k k k k k
  5 10 15 20 5 10 15 20 5 10 15 20 5 10 15 20 5 10 15 20
Cicada 16.1 15.7 15.2 15.3 16.5 16.6 15.6 15.7 15.9 15.7 15.8 15.4 16.4 15.9 15.5 15.3 16.3 16.2 16.2 15.6
  ± 0.3 ± 0.3 ± 0.2 ± 0.2 ± 0.3 ± 0.4 ± 0.2 ± 0.2 ± 0.3 ± 0.3 ± 0.3 ± 0.2 ± 0.4 ± 0.3 ± 0.3 ± 0.2 ± 0.4 ± 0.3 ± 0.3 ± 0.3
Drums 12.5 9.0 11.6 11.1 12.9 9.5 11.8 11.4 11.7 12.1 13.4 14.0 17.9 17.4 20.7 19.9 9.9 10.5 12.7 13.7
  ± 1.7 ± 0.5 ± 0.6 ± 0.4 ± 1.5 ± 0.5 ± 0.6 ± 0.5 ± 1.6 ± 0.7 ± 0.5 ± 0.4 ± 2.1 ± 1.1 ± 0.4 ± 0.4 ± 0.8 ± 0.4 ± 0.4 ± 0.4
Sea 8.1 9.4 10.3 9.8 8.5 10.0 10.5 10.2 5.8 7.6 9.5 9.8 6.6 8.5 9.2 9.3 5.1 6.4 7.5 9.9
  ± 1.8 ± 0.9 ± 0.7 ± 1.0 ± 1.8 ± 0.9 ± 0.7 ± 1.0 ± 1.4 ± 1.3 ± 1.2 ± 1.0 ± 1.5 ± 0.9 ± 0.7 ± 1.1 ± 1.4 ± 1.1 ± 1.1 ± 1.1
Subway 5.0 6.3 8.6 7.1 5.4 6.8 8.8 7.3 5.9 7.1 10.0 9.5 7.8 11.6 13.6 11.9 5.0 6.3 6.8 8.3
  ± 1.3 ± 1.0 ± 0.5 ± 1.0 ± 1.3 ± 1.1 ± 0.5 ± 1.0 ± 1.3 ± 1.2 ± 0.9 ± 1.0 ± 0.9 ± 1.1 ± 0.7 ± 0.5 ± 1.4 ± 1.0 ± 1.0 ± 0.8
  1. The bold figure in each line indicates the best score for a specific noise