Table 2 PSDS values for different TCL lags showing mean and standard deviation. The standard deviations are computed from 5 iterations of each model
From: Multi-rate modulation encoding via unsupervised learning for audio event detection
TCL lags | PSDS1 | PSDS2 |
|---|---|---|
No TCL | 0.375± 0.006 | 0.627 ± 0.005 |
1,1,1 | 0.379 ± 0.008 | 0.641 ± 0.004 |
2,2,2 | 0.376 ± 0.004 | 0.642 \(\varvec{\pm }\) 0.001 |
4,4,4 | 0.379 ± 0.004 | 0.637 ± 0.007 |
8,8,8 | 0.377 ± 0.002 | 0.636 ± 0.006 |
8,4,2 | 0.385 \(\varvec{\pm }\) 0.001 | 0.642 \(\varvec{\pm }\) 0.003 |