she's one of the newer phase connect vtubers from phase saga!

anvuew's room de-reverb model has made the process of making a dataset for this voice model more bearable than it was before bc god damn she has a lot of room reverb.

trained on 15 (and a half) minutes of audio data from the google search checking stream
(voc_fv4 + anvuew's room de-reverb model + dialogue isolate + spectral de-noise + de-click + eq, resampled to 32k)

pitch extraction: rmvpe
steps: 19.8k
batch size: 6
pretrain: klm 4 hfg / 32k

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

Additional Details:

Tags: No tags available

Download Link: https://huggingface.co/sxndypz/rvc-v2-models/resolve/main/nityanil.zip Hosted on Hugging Face