Sweet girl/cute girl/Shaoluo supports singing RVC models

To distinguish the difference before and after the reasoning, I specifically changed it here: the original voice before and after the reasoning uses a male voice, while the output effect is a female voice to make it easier to distinguish.

The training dataset wasn't very large. I originally wanted to train a 48K model, but after training for too long, I ended up training a 44K model, so the difference wasn't significant.

Supports singing, the pitch is pretty good, and I pushed 3-6 songs in a row. Except for some songs with harmonies that aren't very reasonable, the rest is okay, not too much AI flavor.

The right side is the effect area. If you like it, please support me.