Description
Trained by me (). Trained on all Toontown Dog sounds. Best used on a high pitch (e.g., 10-14) for regular male voices. - Dataset Length: 6 seconds - Batch Size: 6 - Crepe Hop Length: 64 - RVC v2
Comments

No way 😂

<:FlippyHappy:672155201939046429>

Why 3000 😂

300 is my go-to number for epochs but for small models like these, I feel like I gotta train it more since it appears barely trained on the *loss/g/total/* graph but idk if that really makes a big difference lol

I just kept training it until it seemed to lower down enough but not to the maximum point before overtraining

gotcha maybe i should try that for the tt cat
3000?
n
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