Speech Separation on LibriCSS Utterance-wise Single-channel (test)
4.2WER (OS)WavLM Large
Evaluation Results
| Method | Links | |||||||||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| WavLM LargePre-trained parameters=frozen, Evaluation ASR model=E2E Transformer [69]2021.10 | 4.2 | — | — | — | — | — | — | — | — | — | — | — | — | 4.1 | 4.8 | 5.8 | 7.4 | 8.5 | 6 | |
| Conformer (rerun)Pre-trained parameters=frozen, Evaluation ASR model=E2E Transformer [69]2021.10 | 4.5 | — | — | — | — | — | — | — | — | — | — | — | — | 4.4 | 6.2 | 8.5 | 11 | 12.6 | 8.3 | |
| WavLM Base+Pre-trained parameters=frozen, Evaluation ASR model=E2E Transformer [69]2021.10 | 4.5 | — | — | — | — | — | — | — | — | — | — | — | — | 4.4 | 5.6 | 7.5 | 9.4 | 10.9 | 7.4 | |
| WavLM Base+Pre-trained parameters=unfrozen, Evaluation ASR model=E2E Transformer [69]2021.10 | 4.5 | — | — | — | — | — | — | — | — | — | — | — | — | 4.3 | 5.9 | 8.3 | 11.1 | 12.5 | 8.2 | |
| HuBERT BasePre-trained parameters=frozen, Evaluation ASR model=E2E Transformer [69]2021.10 | 4.7 | — | — | — | — | — | — | — | — | — | — | — | — | 4.6 | 6.1 | 7.9 | 10.6 | 12.3 | 8.1 | |
| ConformerPre-trained parameters=frozen, Evaluation ASR model=E2E Transformer [69]2021.10 | 5.4 | — | — | — | — | — | — | — | — | — | — | — | — | 5 | 7.5 | 10.7 | 13.8 | 17.1 | 10.6 | |
| BLSTMBackbone=BLSTM, Input Modality=Single-channel2020.08 | — | 15.8 | 6.4 | 14.2 | 5.8 | 18.9 | 9.6 | 25.4 | 15.3 | 31.6 | 20.5 | 35.5 | 25.2 | — | — | — | — | — | — | |
| Conformer-baseBackbone=Conformer, Model Scale=base, Input Modality=Single-channel2020.08 | — | 13.8 | 5.6 | 12.5 | 5.4 | 16.7 | 8.2 | 21.6 | 11.8 | 26.1 | 15.5 | 30.1 | 18.9 | — | — | — | — | — | — | |
| Conformer-largeBackbone=Conformer, Model Scale=large, Input Modality=Single-channel2020.08 | — | 12.9 | 5.4 | 12.2 | 5 | 15.1 | 7.5 | 20.1 | 10.7 | 24.3 | 13.8 | 27.6 | 17.1 | — | — | — | — | — | — | |
| Transformer-baseBackbone=Transformer, Model Scale=base, Input Modality=Single-channel2020.08 | — | 13.2 | 5.5 | 12.3 | 5.2 | 16.5 | 8.3 | 21.8 | 12.1 | 26.2 | 15.6 | 30.6 | 19.3 | — | — | — | — | — | — | |
| Transformer-largeBackbone=Transformer, Model Scale=large, Input Modality=Single-channel2020.08 | — | 13 | 5.3 | 12.4 | 5.1 | 15.5 | 7.4 | 20.1 | 11.1 | 24.6 | 13.5 | 27.9 | 17 | — | — | — | — | — | — |