import gradio as gr
from rvc import *
from argparse import ArgumentParser
import os, sys
if __name__ == '__main__':
parser = ArgumentParser(description='gorge rvc', add_help=True)
parser.add_argument("--share", action="store_true", dest="share_enabled", default=False, help="Enable sharing")
parser.add_argument("--listen", action="store_true", default=False, help="Make the WebUI reachable from your local network.")
parser.add_argument('--listen-host', type=str, help='The hostname that the server will use.')
parser.add_argument('--listen-port', type=int, help='The listening port that the server will use.')
args = parser.parse_args()
warning = sys.argv[0]
if warning == 'True':
warning = True
elif warning == 'False':
warning = False
with gr.Blocks(theme=gr.themes.Base(), title='Mangio-RVC-Web 💻') as app:
gr.HTML(" RVC V2 Huggingface Version
")
if warning:
gr.HTML(" Huggingface version of Easy GUI by Rejekts ")
with gr.Tabs():
with gr.TabItem("Inference"):
# Inference Preset Row
# with gr.Row():
# mangio_preset = gr.Dropdown(label="Inference Preset", choices=sorted(get_presets()))
# mangio_preset_name_save = gr.Textbox(
# label="Your preset name"
# )
# mangio_preset_save_btn = gr.Button('Save Preset', variant="primary")
# Other RVC stuff
with gr.Row():
sid0 = gr.Dropdown(label="1.Choose your Model.", choices=sorted(names), value=check_for_name())
refresh_button = gr.Button("Refresh", variant="primary")
if check_for_name() != '':
get_vc(sorted(names)[0])
vc_transform0 = gr.Number(label="Optional: You can change the pitch here or leave it at 0.", value=0)
#clean_button = gr.Button(i18n("卸载音色省显存"), variant="primary")
spk_item = gr.Slider(
minimum=0,
maximum=2333,
step=1,
label=i18n("请选择说话人id"),
value=0,
visible=False,
interactive=True,
)
#clean_button.click(fn=clean, inputs=[], outputs=[sid0])
sid0.change(
fn=get_vc,
inputs=[sid0],
outputs=[spk_item],
)
but0 = gr.Button("Convert", variant="primary")
with gr.Row():
with gr.Column():
with gr.Row():
dropbox = gr.File(label="Drop your audio here & hit the Reload button.")
with gr.Row():
record_button=gr.Audio(source="microphone", label="OR Record audio.", type="filepath")
with gr.Row():
input_audio0 = gr.Dropdown(
label="2.Choose your audio.",
value="./audios/someguy.mp3",
choices=audio_files
)
dropbox.upload(fn=save_to_wav2, inputs=[dropbox], outputs=[input_audio0])
dropbox.upload(fn=change_choices2, inputs=[], outputs=[input_audio0])
refresh_button2 = gr.Button("Refresh", variant="primary", size='sm')
record_button.change(fn=save_to_wav, inputs=[record_button], outputs=[input_audio0])
record_button.change(fn=change_choices2, inputs=[], outputs=[input_audio0])
with gr.Row():
with gr.Accordion('Text To Speech', open=False):
with gr.Column():
lang = gr.Radio(label='Chinese & Japanese do not work with ElevenLabs currently.',choices=['en','es','fr','pt','zh-CN','de','hi','ja'], value='en')
api_box = gr.Textbox(label="Enter your API Key for ElevenLabs, or leave empty to use GoogleTTS", value='')
elevenid=gr.Dropdown(label="Voice:", choices=eleven_voices)
with gr.Column():
tfs = gr.Textbox(label="Input your Text", interactive=True, value="This is a test.")
tts_button = gr.Button(value="Speak")
tts_button.click(fn=elevenTTS, inputs=[api_box,tfs, elevenid, lang], outputs=[record_button, input_audio0])
with gr.Row():
with gr.Accordion('Wav2Lip', open=False):
with gr.Row():
size = gr.Radio(label='Resolution:',choices=['Half','Full'])
face = gr.UploadButton("Upload A Character",type='file')
faces = gr.Dropdown(label="OR Choose one:", choices=['None','Ben Shapiro','Andrew Tate'])
with gr.Row():
preview = gr.Textbox(label="Status:",interactive=False)
face.upload(fn=success_message,inputs=[face], outputs=[preview, faces])
with gr.Row():
animation = gr.Video(type='filepath')
refresh_button2.click(fn=change_choices2, inputs=[], outputs=[input_audio0, animation])
with gr.Row():
animate_button = gr.Button('Animate')
with gr.Column():
with gr.Accordion("Index Settings", open=False):
file_index1 = gr.Dropdown(
label="3. Path to your added.index file (if it didn't automatically find it.)",
choices=get_indexes(),
value=get_index(),
interactive=True,
)
sid0.change(fn=match_index, inputs=[sid0],outputs=[file_index1])
refresh_button.click(
fn=change_choices, inputs=[], outputs=[sid0, file_index1]
)
# file_big_npy1 = gr.Textbox(
# label=i18n("特征文件路径"),
# value="E:\\codes\py39\\vits_vc_gpu_train\\logs\\mi-test-1key\\total_fea.npy",
# interactive=True,
# )
index_rate1 = gr.Slider(
minimum=0,
maximum=1,
label=i18n("检索特征占比"),
value=0.66,
interactive=True,
)
vc_output2 = gr.Audio(
label="Output Audio (Click on the Three Dots in the Right Corner to Download)",
type='filepath',
interactive=False,
)
animate_button.click(fn=mouth, inputs=[size, face, vc_output2, faces], outputs=[animation, preview])
with gr.Accordion("Advanced Settings", open=False):
f0method0 = gr.Radio(
label="Optional: Change the Pitch Extraction Algorithm.\nExtraction methods are sorted from 'worst quality' to 'best quality'.\nmangio-crepe may or may not be better than rmvpe in cases where 'smoothness' is more important, but rmvpe is the best overall.",
choices=["pm", "dio", "crepe-tiny", "mangio-crepe-tiny", "crepe", "harvest", "mangio-crepe", "rmvpe"], # Fork Feature. Add Crepe-Tiny
value="rmvpe",
interactive=True,
)
crepe_hop_length = gr.Slider(
minimum=1,
maximum=512,
step=1,
label="Mangio-Crepe Hop Length. Higher numbers will reduce the chance of extreme pitch changes but lower numbers will increase accuracy. 64-192 is a good range to experiment with.",
value=120,
interactive=True,
visible=False,
)
f0method0.change(fn=whethercrepeornah, inputs=[f0method0], outputs=[crepe_hop_length])
filter_radius0 = gr.Slider(
minimum=0,
maximum=7,
label=i18n(">=3则使用对harvest音高识别的结果使用中值滤波，数值为滤波半径，使用可以削弱哑音"),
value=3,
step=1,
interactive=True,
)
resample_sr0 = gr.Slider(
minimum=0,
maximum=48000,
label=i18n("后处理重采样至最终采样率，0为不进行重采样"),
value=0,
step=1,
interactive=True,
visible=False
)
rms_mix_rate0 = gr.Slider(
minimum=0,
maximum=1,
label=i18n("输入源音量包络替换输出音量包络融合比例，越靠近1越使用输出包络"),
value=0.21,
interactive=True,
)
protect0 = gr.Slider(
minimum=0,
maximum=0.5,
label=i18n("保护清辅音和呼吸声，防止电音撕裂等artifact，拉满0.5不开启，调低加大保护力度但可能降低索引效果"),
value=0.33,
step=0.01,
interactive=True,
)
formanting = gr.Checkbox(
value=bool(DoFormant),
label="[EXPERIMENTAL] Formant shift inference audio",
info="Used for male to female and vice-versa conversions",
interactive=True,
visible=True,
)
formant_preset = gr.Dropdown(
value='',
choices=get_fshift_presets(),
label="browse presets for formanting",
visible=bool(DoFormant),
)
formant_refresh_button = gr.Button(
value='\U0001f504',
visible=bool(DoFormant),
variant='primary',
)
#formant_refresh_button = ToolButton( elem_id='1')
#create_refresh_button(formant_preset, lambda: {"choices": formant_preset}, "refresh_list_shiftpresets")
qfrency = gr.Slider(
value=Quefrency,
info="Default value is 1.0",
label="Quefrency for formant shifting",
minimum=0.0,
maximum=16.0,
step=0.1,
visible=bool(DoFormant),
interactive=True,
)
tmbre = gr.Slider(
value=Timbre,
info="Default value is 1.0",
label="Timbre for formant shifting",
minimum=0.0,
maximum=16.0,
step=0.1,
visible=bool(DoFormant),
interactive=True,
)
formant_preset.change(fn=preset_apply, inputs=[formant_preset, qfrency, tmbre], outputs=[qfrency, tmbre])
frmntbut = gr.Button("Apply", variant="primary", visible=bool(DoFormant))
formanting.change(fn=formant_enabled,inputs=[formanting,qfrency,tmbre,frmntbut,formant_preset,formant_refresh_button],outputs=[formanting,qfrency,tmbre,frmntbut,formant_preset,formant_refresh_button])
frmntbut.click(fn=formant_apply,inputs=[qfrency, tmbre], outputs=[qfrency, tmbre])
formant_refresh_button.click(fn=update_fshift_presets,inputs=[formant_preset, qfrency, tmbre],outputs=[formant_preset, qfrency, tmbre])
with gr.Row():
vc_output1 = gr.Textbox("")
f0_file = gr.File(label=i18n("F0曲线文件, 可选, 一行一个音高, 代替默认F0及升降调"), visible=False)
but0.click(
vc_single,
[
spk_item,
input_audio0,
vc_transform0,
f0_file,
f0method0,
file_index1,
# file_index2,
# file_big_npy1,
index_rate1,
filter_radius0,
resample_sr0,
rms_mix_rate0,
protect0,
crepe_hop_length
],
[vc_output1, vc_output2],
)
with gr.Accordion("Batch Conversion",open=False):
with gr.Row():
with gr.Column():
vc_transform1 = gr.Number(
label=i18n("变调(整数, 半音数量, 升八度12降八度-12)"), value=0
)
opt_input = gr.Textbox(label=i18n("指定输出文件夹"), value="opt")
f0method1 = gr.Radio(
label=i18n(
"选择音高提取算法,输入歌声可用pm提速,harvest低音好但巨慢无比,crepe效果好但吃GPU"
),
choices=["pm", "harvest", "crepe", "rmvpe"],
value="rmvpe",
interactive=True,
)
filter_radius1 = gr.Slider(
minimum=0,
maximum=7,
label=i18n(">=3则使用对harvest音高识别的结果使用中值滤波，数值为滤波半径，使用可以削弱哑音"),
value=3,
step=1,
interactive=True,
)
with gr.Column():
file_index3 = gr.Textbox(
label=i18n("特征检索库文件路径,为空则使用下拉的选择结果"),
value="",
interactive=True,
)
file_index4 = gr.Dropdown(
label=i18n("自动检测index路径,下拉式选择(dropdown)"),
choices=sorted(index_paths),
interactive=True,
)
refresh_button.click(
fn=lambda: change_choices()[1],
inputs=[],
outputs=file_index4,
)
# file_big_npy2 = gr.Textbox(
# label=i18n("特征文件路径"),
# value="E:\\codes\\py39\\vits_vc_gpu_train\\logs\\mi-test-1key\\total_fea.npy",
# interactive=True,
# )
index_rate2 = gr.Slider(
minimum=0,
maximum=1,
label=i18n("检索特征占比"),
value=1,
interactive=True,
)
with gr.Column():
resample_sr1 = gr.Slider(
minimum=0,
maximum=48000,
label=i18n("后处理重采样至最终采样率，0为不进行重采样"),
value=0,
step=1,
interactive=True,
)
rms_mix_rate1 = gr.Slider(
minimum=0,
maximum=1,
label=i18n("输入源音量包络替换输出音量包络融合比例，越靠近1越使用输出包络"),
value=1,
interactive=True,
)
protect1 = gr.Slider(
minimum=0,
maximum=0.5,
label=i18n(
"保护清辅音和呼吸声，防止电音撕裂等artifact，拉满0.5不开启，调低加大保护力度但可能降低索引效果"
),
value=0.33,
step=0.01,
interactive=True,
)
with gr.Column():
dir_input = gr.Textbox(
label=i18n("输入待处理音频文件夹路径(去文件管理器地址栏拷就行了)"),
value="E:\codes\py39\\test-20230416b\\todo-songs",
)
inputs = gr.File(
file_count="multiple", label=i18n("也可批量输入音频文件, 二选一, 优先读文件夹")
)
with gr.Row():
format1 = gr.Radio(
label=i18n("导出文件格式"),
choices=["wav", "flac", "mp3", "m4a"],
value="flac",
interactive=True,
)
but1 = gr.Button(i18n("转换"), variant="primary")
vc_output3 = gr.Textbox(label=i18n("输出信息"))
but1.click(
vc_multi,
[
spk_item,
dir_input,
opt_input,
inputs,
vc_transform1,
f0method1,
file_index3,
file_index4,
# file_big_npy2,
index_rate2,
filter_radius1,
resample_sr1,
rms_mix_rate1,
protect1,
format1,
crepe_hop_length,
],
[vc_output3],
)
but1.click(fn=lambda: easy_uploader.clear())
with gr.TabItem("Download Model"):
with gr.Row():
url=gr.Textbox(label="Enter the URL to the Model:")
with gr.Row():
model = gr.Textbox(label="Name your model:")
download_button=gr.Button("Download")
with gr.Row():
status_bar=gr.Textbox(label="")
download_button.click(fn=download_from_url, inputs=[url, model], outputs=[status_bar])
with gr.Row():
gr.Markdown(
"""
Original RVC:https://github.com/RVC-Project/Retrieval-based-Voice-Conversion-WebUI Mangio’s RVC Fork:https://github.com/Mangio621/Mangio-RVC-Fork ❤️ If you like the EasyGUI, help me keep it.❤️ https://paypal.me/lesantillan
"""
)
app.launch(share=args.share_enabled)