OPENGIT/Retrieval-based-Voice-Conversion-WebUI
1
0
Fork 0
mirror of https://github.com/RVC-Project/Retrieval-based-Voice-Conversion-WebUI.git synced 2025-01-31 02:32:51 +08:00
Code Issues Packages Projects Releases Wiki Activity

优化print

Browse Source
This commit is contained in:
源文雨 2023-04-10 18:34:10 +08:00
parent ff1a711cad
commit 80bc765cbc
2 changed files with 1 additions and 6 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

2
infer-web.py
View File

@ -79,7 +79,7 @@ def vc_single(sid,input_audio,f0_up_key,f0_file,f0_method,file_index,file_big_np
if(hubert_model==None):load_hubert() if(hubert_model==None):load_hubert()
if_f0 = cpt.get("f0", 1) if_f0 = cpt.get("f0", 1)
audio_opt=vc.pipeline(hubert_model,net_g,sid,audio,times,f0_up_key,f0_method,file_index,file_big_npy,index_rate,if_f0,f0_file=f0_file) audio_opt=vc.pipeline(hubert_model,net_g,sid,audio,times,f0_up_key,f0_method,file_index,file_big_npy,index_rate,if_f0,f0_file=f0_file)
print(times) print("npy: ", times[0], "s, f0:", times[1], "s, infer: ", times[2], "s", sep='')
return "Success", (tgt_sr, audio_opt) return "Success", (tgt_sr, audio_opt)
except: except:
info=traceback.format_exc() info=traceback.format_exc()

5
vc_infer_pipeline.py
View File

@ -72,7 +72,6 @@ class VC(object):
"output_layer": 9, # layer 9 "output_layer": 9, # layer 9
} }
t0 = ttime() t0 = ttime()
print("vc npy start time:", t0)
with torch.no_grad(): with torch.no_grad():
logits = model.extract_features(**inputs) logits = model.extract_features(**inputs)
feats = model.final_proj(logits[0]) feats = model.final_proj(logits[0])
@ -87,7 +86,6 @@ 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)
t1 = ttime() t1 = ttime()
print("vc infer start time:", t1)
p_len = audio0.shape[0]//self.window p_len = audio0.shape[0]//self.window
if(feats.shape[1]<p_len): if(feats.shape[1]<p_len):
p_len=feats.shape[1] p_len=feats.shape[1]
@ -103,7 +101,6 @@ class VC(object):
del feats,p_len,padding_mask del feats,p_len,padding_mask
if torch.cuda.is_available(): torch.cuda.empty_cache() if torch.cuda.is_available(): torch.cuda.empty_cache()
t2 = ttime() t2 = ttime()
print("vc infer end time:", t2)
times[0] += (t1 - t0) times[0] += (t1 - t0)
times[2] += (t2 - t1) times[2] += (t2 - t1)
return audio1 return audio1
@ -128,7 +125,6 @@ class VC(object):
audio_opt=[] audio_opt=[]
t=None t=None
t1=ttime() t1=ttime()
print("f0 start time:", t1)
audio_pad = np.pad(audio, (self.t_pad, self.t_pad), mode='reflect') audio_pad = np.pad(audio, (self.t_pad, self.t_pad), mode='reflect')
p_len=audio_pad.shape[0]//self.window p_len=audio_pad.shape[0]//self.window
inp_f0=None inp_f0=None
@ -150,7 +146,6 @@ class VC(object):
pitch = torch.tensor(pitch,device=self.device).unsqueeze(0).long() pitch = torch.tensor(pitch,device=self.device).unsqueeze(0).long()
pitchf = torch.tensor(pitchf,device=self.device).unsqueeze(0).float() pitchf = torch.tensor(pitchf,device=self.device).unsqueeze(0).float()
t2=ttime() t2=ttime()
print("f0 end time:", t2)
times[1] += (t2 - t1) times[1] += (t2 - t1)
for t in opt_ts: for t in opt_ts:
t=t//self.window*self.window t=t//self.window*self.window