rvc-infer 2024.1.5.1

Python wrapper for inference with rvc

RVC INFER

---

A streamlined Python wrapper for inference with RVC. Specifically designed for inference tasks.

---

how to install

pip install rvc_infer

How to Use

download online models

from rvc_infer import download_online_model



output = download_online_model(
    url,
    dir_name
)


print(output)

Infernece

from rvc_infer import infer_audio


result = infer_audio(
    model_name,
    audio_path,
    f0_change=0,
    f0_method="rmvpe+",
    min_pitch="50",
    max_pitch="1100",
    crepe_hop_length=128,
    index_rate=0.75,
    filter_radius=3,
    rms_mix_rate=0.25,
    protect=0.33,
    split_infer=False,
    min_silence=500,
    silence_threshold=-50,
    seek_step=1,
    keep_silence=100,
    do_formant=False,
    quefrency=0,
    timbre=1,
    f0_autotune=False,
    audio_format="wav",
    resample_sr=0,
    hubert_model_path="hubert_base.pt",
    rmvpe_model_path="rmvpe.pt",
    fcpe_model_path="fcpe.pt"
    )

little note

the model_name It will automaticly search a folder containing the pth file and index file.