Voice model of Jeanne Alter from Fate/Grand Order

V1 Voice model was trained with Mangio-Crepe, using clips from FGO only.

V2 Voice model was trained with Mangio-Crepe, using clips from FGO and FGO Arcade.

Model Details

Update 2024/1/4: The voice has received an upgrade, but for some reason... the V2 sounds a bit different? So pick the one you prefer.

I attached two files to the V2 since the training graph was a bit odd, and I couldn't tell the difference between the 100 epoch and 300 epoch. Please raise a pull request if you have a strong opinion about it.

Please use RMVPE when sampling with speaking clips that involve some distortion such as reverb, even though the model was trained with Mangio-Crepe.

When singing, it's better to use Mangio-Crepe.

With other clips, I can't tell a difference between RMVPE and Mangio-Crepe.

Warnings

They may not come out as amazing as they should, so extra tweaking may be necessary.

Jalter has a very high voice, and using it with lower voice models may not sound correct without pitch correction.