yay an another phase connect model after months

trained on 17 minutes of audio data from the rating viewers' desks stream
(mel-roformer fv7z + eq + de-plosive + de-click + de-ess, resampled to 32k)

pitch extraction: rmvpe
steps: 18.8k
batch size: 7
pretrain: klm 4 hfg / 32k

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

Additional Details:

Tags: No tags available

Download Link: https://huggingface.co/sxndypz/rvc-v2-models/resolve/main/marimari.zip