No background noise/throat-clearing/ridiculous 48K AI Manxi RVC model
Inference Comparison
After Inference
Other Inference
Model Introduction
The effect can be auditioned for real-time reasoning results.
This is a noise-free RVC model, with a tone leaning towards a mature female tone, depending on the original input timbre.
After multiple rounds of testing, the clarity and smiling are quite stable, and it also supports accent performance well, so there won't be any changes in tone or off-key when the accent is struck.
However, this model still does not support screaming. Due to its limited vocal range, it already adapts well to various tonal expressions.
Currently, the tone used for testing is 12, which can be adjusted according to your own needs.
Also, pay attention to the harvest parameter. Please adjust the settings according to your own configuration, otherwise you may encounter "lag" or similar issues.
This model is trained using the Mygf-f048k base model, which is especially clear in enunciation and rarely appears unclear.
In short, this is a great mature voice model.
zhangzhang
Version Details
Recommended Parameters
Related Models
Keywords
Comments
No comments yet

