Brand new (Poetic Drama) Mambo 48K RVC model

This is a new Mambo model trained on the Mygf-f048k base model, truly achieving smooth conversion between Chinese and Japanese languages.

The original dataset was entirely Mambo's Japanese dataset and was not converted to Chinese through TTS for training. It retains the Mambo tone and supports multiple languages. There will be no "Colonel sound."

For inference settings, we recommend setting it to 9, 12, or even higher for real-time reasoning, and input in a cute tone.

Let Mambo truly support speaking Chinese without changing his tone. (The sound after reasoning is the result of real-time inference.)

And judging from the detective singing "Qilin," there is absolutely no "strange sound."

For singing, we recommend using the original voice to output the pitch. For some deeply emotional songs, it's suggested to pull the pitch a bit more.

If you have purchased before

https://apis.klrvc.com/mxgf/1990.html

For this model, please find the purchase order number in " My Orders" and send me a private message (if I don't reply in time and the site owner ships on my behalf), I will send you the latest Mambo model for free.