Girly and mature older sister, 48k sampling rate, monthly model

It uses 48K high-sampling corpora, giving it a rather unique tonal character.

This model supports limited emotional changes such as laughter.

This model uses the RMVPE algorithm, but it is recommended to use FCPE for transcription. Other parameters can be set according to the model parameters.

Adjust the pitch to the appropriate position according to each individual's original input.

Overall, the model performs quite well. I recommend testing and adapting it immediately based on the sound system after getting it.