Description
I used 1 minute and 50 seconds of isolated in game lines to train this model and looped it back a fair bit so that it lasted 15 minutes, not sure if that improves it since I'm still learning on how to improve low data set models. Pitch extracted with mangio-crepe. To be quite honest, it still sounds like Pre-Reworked Mundo at most given points during use, just sometimes it will sound less accurate since it's such a limited data pool I was left with. Model
Comments
No comments yet. Start the conversation!
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