Description
< Trained off of cleaned studio acapellas for every song off of Smithereens, which were seperated with kareoke models to give references for both high and low voice Joji, overall making 20~ minutes of data
Comments

litsa_the_dancer
is this rmvpe?

shadowtb
yeah

litsa_the_dancer
wow it did a great job capturing his low vocals

litsa_the_dancer
what about some higher vocals

litsa_the_dancer
how do they sound?

shadowtb
honestly thats the only test I've done I was pushing 4 hours in colab 😭

litsa_the_dancer
relatable

shadowtb
The model has pretty much every vocal range in smithereens

shadowtb
So I'd imagine fine

litsa_the_dancer
Alright that's good to hear. nice model overall 👍
Add a comment
Samples
New
Classic
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