Automatic Speech Recognition on CHiME-4 Simu
4.48WER (Whisper)WER-OA (Eq. 2)
Evaluation results
| Method | Links | |||
|---|---|---|---|---|
| WER-OA (Eq. 2)Speech Enhancement Model=GR-KAN-based MPSENet2026.02 | 4.48 | 4.19 | 15.5 | |
| Conf-OA (Eq. 3)Speech Enhancement Model=GR-KAN-based MPSENet2026.02 | 4.97 | 4.78 | 16.73 | |
| Classifier-OA3classSpeech Enhancement Model=GR-KAN-based MPSENet2026.02 | 5.06 | 4.88 | 16.9 | |
| DNSMOS-OASpeech Enhancement Model=GR-KAN-based MPSENet2026.02 | 5.27 | 5 | 17.09 | |
| Noisy ySpeech Enhancement Model=None2026.02 | 5.36 | 5.24 | 25.82 | |
| Conf-Switch (Eq. 6)Speech Enhancement Model=GR-KAN-based MPSENet2026.02 | 5.93 | 5.32 | 19.48 | |
| Classifier-OA2classSpeech Enhancement Model=GR-KAN-based MPSENet2026.02 | 7.12 | 6.76 | 18.59 | |
| Enhanced x̂Speech Enhancement Model=GR-KAN-based MPSENet2026.02 | 8 | 7.91 | 20.7 |