Table 10 Statistical significance of the MOS results of mono-lingual conversion in seen-to-seen scenario
From: U2-VC: one-shot voice conversion using two-level nested U-structure
| Statistical significance of MOS (similarity) | Statistical significance of MOS (naturalness) | ||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|
| SF2TF | SF2TM | SM2TF | SM2TM | Overall | SF2TF | SF2TM | SM2TF | SM2TM | Overall | ||
| AdaIN-VC | AGAIN-VC | 0.000 | 0.015 | 0.001 | 0.000 | 0.000 | 0.000 | 0.000 | 0.001 | 0.000 | 0.000 |
| U2-VC | 0.001 | 0.000 | 0.000 | 0.000 | 0.000 | 0.000 | 0.000 | 0.000 | 0.000 | 0.000 | |
| AGAIN-VC | AdaIN-VC | 0.000 | 0.015 | 0.001 | 0.000 | 0.000 | 0.000 | 0.000 | 0.001 | 0.000 | 0.000 |
| U2-VC | 0.380 | 0.094 | 0.083 | 0.007 | 0.003 | 0.086 | 0.014 | 0.018 | 0.038 | 0.000 | |
| U2-VC | AdaIN-VC | 0.001 | 0.000 | 0.000 | 0.000 | 0.000 | 0.000 | 0.000 | 0.000 | 0.000 | 0.000 |
| AGAIN-VC | 0.380 | 0.094 | 0.083 | 0.007 | 0.003 | 0.086 | 0.014 | 0.018 | 0.038 | 0.000 | |