Description
trained with all 21 minutes of audio in the original Half-life. there may be some background noise, but the noise in my tests was very minor, and could easily be masked by music or removed in an audio/video editor. you shouldn't need to change the pitch when using crepe unless the reference audio is super high or low pitched, although high pitch voices are harder for it.
Comments
Big Mitch
I already did this model with mangio-crepe and a 35 minute dataset at 800 epoch
Muhammad Nawawi Idris
Gordon Why are you wearing that Ridiculous Ties?! The administrator said the ties are banned
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 models by
More to explore
Loading more