| SLUE: New Benchmark Tasks for Spoken Language Understanding Evaluation on Natural Speech | ✓ Link | 69.6 | 82.2 | DeBERTa-L | W2V2-L-LL60K (pipeline approach, uses LM) | 2021-11-19 |
| SLUE: New Benchmark Tasks for Spoken Language Understanding Evaluation on Natural Speech | ✓ Link | 68.0 | 79.8 | DeBERTa-L | W2V2-B-LS960 (pipeline approach, uses LM) | 2021-11-19 |
| Wav2Seq: Pre-training Speech-to-Text Encoder-Decoder Models Using Pseudo Languages | ✓ Link | 65.4 | | | Wav2Seq (from HuBERT-large) | 2022-05-02 |
| SLUE: New Benchmark Tasks for Spoken Language Understanding Evaluation on Natural Speech | ✓ Link | 64.8 | 73.3 | N/A | W2V2-L-LL60K (e2e approach, uses LM) | 2021-11-19 |
| SLUE: New Benchmark Tasks for Spoken Language Understanding Evaluation on Natural Speech | ✓ Link | 63.4 | 71.7 | N/A | W2V2-B-LS960 (e2e approach, uses LM) | 2021-11-19 |
| SLUE: New Benchmark Tasks for Spoken Language Understanding Evaluation on Natural Speech | ✓ Link | 61.9 | 70.3 | N/A | HuBERT-B-LS960 (e2e approach, uses LM) | 2021-11-19 |
| SLUE: New Benchmark Tasks for Spoken Language Understanding Evaluation on Natural Speech | ✓ Link | 61.8 | 69.8 | N/A | W2V2-B-VP100K (e2e approach, uses LM) | 2021-11-19 |
| SLUE: New Benchmark Tasks for Spoken Language Understanding Evaluation on Natural Speech | ✓ Link | 57.8 | 78.8 | DeBERTa-L | W2V2-L-LL60K (pipeline approach) | 2021-11-19 |
| SLUE: New Benchmark Tasks for Spoken Language Understanding Evaluation on Natural Speech | ✓ Link | 50.9 | 64.7 | - | W2V2-L-LL60K (e2e approach) | 2021-11-19 |
| SLUE: New Benchmark Tasks for Spoken Language Understanding Evaluation on Natural Speech | ✓ Link | 50.2 | 64.0 | - | W2V2-B-LS960 (e2e approach) | 2021-11-19 |
| SLUE: New Benchmark Tasks for Spoken Language Understanding Evaluation on Natural Speech | ✓ Link | 49.8 | 62.9 | - | HuBERT-B-LS960 (e2e approach) | 2021-11-19 |
| SLUE: New Benchmark Tasks for Spoken Language Understanding Evaluation on Natural Speech | ✓ Link | 49.5 | 74.2 | DeBERTa-L | W2V2-B-LS960 (pipeline approach) | 2021-11-19 |
| SLUE: New Benchmark Tasks for Spoken Language Understanding Evaluation on Natural Speech | ✓ Link | 47.9 | 60.8 | - | W2V2-B-VP100K (e2e approach) | 2021-11-19 |