Automatic Speech Recognition on LibriSpeech clean + outdoor noise unseen (test)
9.06WERSSL-SoftDTW
Evaluation results
| Method | Links | |
|---|---|---|
| SSL-SoftDTWSSL Fine-tuning of SE Model=SSL-SoftDTW, Enhancement=Yes2026.01 | 9.06 | |
| SSL-MSE-PADSSL Fine-tuning of SE Model=SSL-MSE-PAD, Enhancement=Yes2026.01 | 9.11 | |
| SSL-MSESSL Fine-tuning of SE Model=SSL-MSE, Enhancement=Yes2026.01 | 9.19 | |
| master641SSL Fine-tuning of SE Model=None, Enhancement=Yes2026.01 | 9.93 | |
| No EnhancementSSL Fine-tuning of SE Model=None, Enhancement=No2026.01 | 12.47 |