Daitaku Helios (Umamusume) (Singing) (RVC v2) - 600 Epochs

[bold pelican]

-# same horse except she sings

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

trained on 8(?) minutes of her singing data extracted from the game (thx maki ligon for the dataset)
(anvuew de-reverb + uvr de-echo (in some parts) + some eq, resampled to 40k)

CV: Aya Yamane

pitch extraction: rmvpe
steps: 21k
batch size: 6
pretrain: klm 4 hfg / 40k

v2: #1462055746513670144

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

(almost forgot to mention these samples are inferred in applio so the inferencing results might differ with whatever program you use to infer audios with)

[gleaming fable]

Model Ready ✅

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

[bold pelican]

just added new data to the singing model and now the data is 20 minutes long

it doesn't strain as much on higher notes unlike the v1 version which is a good thing