Skip to main content

Table 12 Objective evaluation results of voice conversion in cross-lingual scenario

From: U2-VC: one-shot voice conversion using two-level nested U-structure

  Predicted MOS by NISQA
  VCTK2VCC VCC2VCTK VCC2VCC Average
AdaIN-VC 2.83 2.72 2.81 2.79
AGAIN-VC 3.56 3.40 3.64 3.53
U2-VC 3.60 3.82 3.72 3.71