Table 7 Evaluation of Nevisa and the robustness methods on FANOS noisy tasks (WER% on word level)
From: A large vocabulary continuous speech recognition system for Persian language
Robustness | Task A | Task B | Task C | Task D |
|---|---|---|---|---|
None | 74.04 | 75.32 | 116.41 | 105.94 |
VTLN+MLLR | 30.37 | 32.87 | 82.52 | 60.07 |
PMC-MAP | 38.63 | 50.49 | 69.36 | 50.22 |
PC-PMC+MLLR | 31.33 | 28.70 | 56.17 | 42.11 |