Description
softgirl rmvpe sample rate: 32k izotope nectar 4, ozone 11, audacity gptsovits for tts dataset idk im lazy old model
Comments
nuu is just rvc but i made dataset with gpt sovits

are you using regular pretrain?
original pretrain yes

Sounds too damn good to have just 8 mins of dataset.

true

Maybe on some type of voices, with just some minutes of dataset (even less than 10 minutes) you can make a nice model for realtime.

the dataset is very very very very very denoised so um that may be why

RX10 makes miracles.
The most soft spoken model I've ever heard.
Add a comment
Samples
1. Singing
Male
English
2. Singing
Female
English
3. Singing (Dry)
Female
English
4. Singing (High)
Female
English
5. Singing 2
Male
English
6. Singing (Dry)
Male
English
7. Singing (Dry, High)
Male
English
Pitch
More to explore
Loading more