OPENGIT/Retrieval-based-Voice-Conversion-WebUI
1
0
Fork 0
mirror of https://github.com/RVC-Project/Retrieval-based-Voice-Conversion-WebUI.git synced 2025-01-06 15:05:05 +08:00
Code Issues Packages Projects Releases Wiki Activity

fix-no-f0-model-protect-issue

Browse Source 
fix-no-f0-model-protect-issue
This commit is contained in:
RVC-Boss 2023-06-18 15:17:36 +08:00 committed by GitHub
parent ec0c39d9bc
commit a9a77f2556
No known key found for this signature in database
GPG Key ID: 4AEE18F83AFDEB23
2 changed files with 16 additions and 11 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

21
infer-web.py
View File

@ -3,7 +3,7 @@ import shutil
import sys import sys
now_dir = os.getcwd() now_dir = os.getcwd()
sys.path.append(now_dir) sys.path.append(now_dir)
import traceback import traceback,pdb
import warnings import warnings
import numpy as np import numpy as np
@ -396,7 +396,7 @@ def uvr(model_name, inp_root, save_root_vocal, paths, save_root_ins, agg, format
# 一个选项卡全局只能有一个音色 # 一个选项卡全局只能有一个音色
def get_vc(sid): def get_vc(sid,to_return_protect0,to_return_protect1):
global n_spk, tgt_sr, net_g, vc, cpt, version global n_spk, tgt_sr, net_g, vc, cpt, version
if sid == "" or sid == []: if sid == "" or sid == []:
global hubert_model global hubert_model
@ -434,6 +434,11 @@ def get_vc(sid):
tgt_sr = cpt["config"][-1] tgt_sr = cpt["config"][-1]
cpt["config"][-3] = cpt["weight"]["emb_g.weight"].shape[0] # n_spk cpt["config"][-3] = cpt["weight"]["emb_g.weight"].shape[0] # n_spk
if_f0 = cpt.get("f0", 1) if_f0 = cpt.get("f0", 1)
if(if_f0==0):
to_return_protect0=to_return_protect1={"visible": False, "value": 0.5, "__type__": "update"}
else:
to_return_protect0 ={"visible": True, "value": to_return_protect0, "__type__": "update"}
to_return_protect1 ={"visible": True, "value": to_return_protect1, "__type__": "update"}
version = cpt.get("version", "v1") version = cpt.get("version", "v1")
if version == "v1": if version == "v1":
if if_f0 == 1: if if_f0 == 1:
@ -454,7 +459,7 @@ def get_vc(sid):
net_g = net_g.float() net_g = net_g.float()
vc = VC(tgt_sr, config) vc = VC(tgt_sr, config)
n_spk = cpt["config"][-3] n_spk = cpt["config"][-3]
return {"visible": True, "maximum": n_spk, "__type__": "update"} return {"visible": True, "maximum": n_spk, "__type__": "update"},to_return_protect0,to_return_protect1
def change_choices(): def change_choices():
@ -1247,11 +1252,6 @@ with gr.Blocks() as app:
interactive=True, interactive=True,
) )
clean_button.click(fn=clean, inputs=[], outputs=[sid0]) clean_button.click(fn=clean, inputs=[], outputs=[sid0])
sid0.change(
fn=get_vc,
inputs=[sid0],
outputs=[spk_item],
)
with gr.Group(): with gr.Group():
gr.Markdown( gr.Markdown(
value=i18n("男转女推荐+12key, 女转男推荐-12key, 如果音域爆炸导致音色失真也可以自己调整到合适音域. ") value=i18n("男转女推荐+12key, 女转男推荐-12key, 如果音域爆炸导致音色失真也可以自己调整到合适音域. ")
@ -1475,6 +1475,11 @@ with gr.Blocks() as app:
], ],
[vc_output3], [vc_output3],
) )
sid0.change(
fn=get_vc,
inputs=[sid0,protect0,protect1],
outputs=[spk_item,protect0,protect1],
)
with gr.TabItem(i18n("伴奏人声分离&去混响&去回声")): with gr.TabItem(i18n("伴奏人声分离&去混响&去回声")):
with gr.Group(): with gr.Group():
gr.Markdown( gr.Markdown(

6
vc_infer_pipeline.py
View File

@ -184,7 +184,7 @@ class VC(object):
with torch.no_grad(): with torch.no_grad():
logits = model.extract_features(**inputs) logits = model.extract_features(**inputs)
feats = model.final_proj(logits[0]) if version == "v1" else logits[0] feats = model.final_proj(logits[0]) if version == "v1" else logits[0]
if protect < 0.5: if protect < 0.5 and pitch!=None and pitchf!=None:
feats0 = feats.clone() feats0 = feats.clone()
if ( if (
isinstance(index, type(None)) == False isinstance(index, type(None)) == False
@ -211,7 +211,7 @@ class VC(object):
) )
feats = F.interpolate(feats.permute(0, 2, 1), scale_factor=2).permute(0, 2, 1) feats = F.interpolate(feats.permute(0, 2, 1), scale_factor=2).permute(0, 2, 1)
if protect < 0.5: if protect < 0.5 and pitch!=None and pitchf!=None:
feats0 = F.interpolate(feats0.permute(0, 2, 1), scale_factor=2).permute( feats0 = F.interpolate(feats0.permute(0, 2, 1), scale_factor=2).permute(
0, 2, 1 0, 2, 1
) )
@ -223,7 +223,7 @@ class VC(object):
pitch = pitch[:, :p_len] pitch = pitch[:, :p_len]
pitchf = pitchf[:, :p_len] pitchf = pitchf[:, :p_len]
if protect < 0.5: if protect < 0.5 and pitch!=None and pitchf!=None:
pitchff = pitchf.clone() pitchff = pitchf.clone()
pitchff[pitchf > 0] = 1 pitchff[pitchf > 0] = 1
pitchff[pitchf < 1] = protect pitchff[pitchf < 1] = protect