| Task Name | Dataset Name | SOTA Result | Trend | |
|---|---|---|---|---|
| Naturalness | CMU ARCTIC | UTMOS4.33 | 39 | |
| Intelligibility | CMU ARCTIC | WER2.61 | 39 | |
| Speaker Similarity | CMU ARCTIC | EER0.62 | 39 | |
| Voice Conversion | CMU ARCTIC (test) | Speaker Similarity2.97 | 39 | |
| Blind Source Separation | CMU ARCTIC Noisy condition SNR uniformly sampled from [15, 25] dB | SDRi (dB)3.41 | 24 | |
| Blind Source Separation | CMU ARCTIC Noiseless condition SNR = ∞ | SDRi (dB)4.9 | 24 | |
| Voice Conversion | CMU Arctic clb to slt | MCD5.87 | 5 | |
| Voice Conversion | CMU Arctic bdl to slt | MCD5.93 | 5 | |
| Vocoding | CMU-ARCTIC UNS-USQ (test) | PESQ1.8063 | 2 |