Marvelous Sunday (Singing) (Umamusume) (RVC v2) - 480 Epochs

[noble coyote]

marvelous sunday's here with the marvelous singing capabilities!!!
-# (woohoo no one cares)

this character's featured in umamusume (both the anime and game)

trained on 10 minutes of her singing data extracted from the game (thx maki ligon for the dataset)
(anvuew de-reverb + sucial de-reverb/de-echo (in some parts) + uvr de-echo (in some parts), resampled to 40k)

CV: Marie Miyake

pitch extraction: rmvpe
steps: 19.6k
batch size: 6
pretrain: klm 4 hfg / 40k

huggingface: https://huggingface.co/sxndypz/rvc-v2-models/resolve/main/marveloussunday_sing.zip

(all of the samples above are inferred in applio, so the results from inferencing might differ)

[proven pasture]

Model Ready ✅

This model has been synced with Weights and is ready to use for free!

[nimble spruce]

how'd u get the voice lines o - o

[noble coyote]

they've been extracted using umamusumeexplorer i believe

either i extract it or someone extracts the lines for me (if i don't have the characters i need to make models out of bc global yk)

[latent perch]

yeah im using that with the japanese version of the game

[nimble spruce]

same im also using the japanese ver of the game

[noble coyote]

new sample

[weak oar]

could you send me the sample you used for talking and singing @noble coyote ?