Ben Bigger (Current Patch) | ZenlessZoneZero | 32k | RVCv2 | 623 Epochs

[strong socket]

download link: https://huggingface.co/lyery/mymodels/resolve/main/bear.zip?download=true

hes cute
what can i say man, anyways, this model can sing because despite the dataset being super short, his voicelines are expressive enough for rvc + i had to use batch 16 because i ran out of money on colab lmao

Info:
Sample rate: 32k
Pitch e: rmvpe
Batch size: 16
Pretrain: Original
Dataset length: 2:27 minutes
Precision:Fp32
Epoch: best g/total tho best kl is actually a few epochs earlier, but ya know g/total > everything (in reality this one has slightly improved sibilances vs best kl, those are always gonna be shit in smol datasets xD)

[wild swallow]

Rate the model

To rate this model, either use /rate or click the buttons below.

[strong socket]

i added the download link in the embedded files somehow wtf discord

[night blade]

Now we got the two simp characters out of God knows how many

[sonic tartan]

you should add the ''resolve'' link of your model, not the ''blob'' link xD

https://huggingface.co/lyery/mymodels/resolve/main/bear.zip?download=true

here is the corrected link

[strong socket]

Thx lols i was so tired when i made this post that i didn’t notice the link was wrong

[sonic tartan]

np

[mortal jasper]

Model Ready ✅

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