From: AUC optimization for deep learning-based voice activity detection
Noise type | SNR | MCE | MMSE | MaxAUCsigm | MaxAUChinge |
---|---|---|---|---|---|
Babble | − 10 dB | 0.5752 | 0.5664 | 0.5833 | 0.5799 |
− 5 dB | 0.6442 | 0.6391 | 0.6588 | 0.6565 | |
0 dB | 0.7272 | 0.7222 | 0.7462 | 0.7441 | |
5 dB | 0.7900 | 0.7867 | 0.8076 | 0.8057 | |
10 dB | 0.8246 | 0.8289 | 0.8387 | 0.8390 | |
15 dB | 0.8420 | 0.8430 | 0.8529 | 0.8467 | |
20 dB | 0.8487 | 0.8624 | 0.8579 | 0.8628 | |
Factory | − 10 dB | 0.5992 | 0.5938 | 0.6115 | 0.6011 |
− 5 dB | 0.6743 | 0.6694 | 0.6897 | 0.6822 | |
0 dB | 0.7340 | 0.7294 | 0.7536 | 0.7474 | |
5 dB | 0.7791 | 0.7769 | 0.7994 | 0.7929 | |
10 dB | 0.8142 | 0.8166 | 0.8299 | 0.8285 | |
15 dB | 0.8373 | 0.8458 | 0.8485 | 0.8503 | |
20 dB | 0.8474 | 0.8603 | 0.8569 | 0.8646 | |
Volvo | − 10 dB | 0.7571 | 0.7551 | 0.7790 | 0.7858 |
− 5 dB | 0.7933 | 0.7905 | 0.8195 | 0.8270 | |
0 dB | 0.8244 | 0.8229 | 0.8443 | 0.8534 | |
5 dB | 0.8350 | 0.8343 | 0.8530 | 0.8602 | |
10 dB | 0.8374 | 0.8295 | 0.8560 | 0.8602 | |
15 dB | 0.8423 | 0.8518 | 0.8600 | 0.8589 | |
20 dB | 0.8476 | 0.8595 | 0.8645 | 0.8593 |