Merge branch 'RVC-Boss:main' into main

GPT_SoVITS/prepare_datasets/2-get-hubert-wav32k.py

@@ -49,10 +49,13 @@ maxx=0.95
 alpha=0.5
 device="cuda:0"
 model=cnhubert.get_model()
+# is_half=False
 if(is_half==True):
     model=model.half().to(device)
 else:
     model = model.to(device)
+
+nan_fails=[]
 def name2go(wav_name):
     hubert_path="%s/%s.pt"%(hubert_dir,wav_name)
     if(os.path.exists(hubert_path)):return
@@ -60,25 +63,27 @@ def name2go(wav_name):
     tmp_audio = load_audio(wav_path, 32000)
     tmp_max = np.abs(tmp_audio).max()
     if tmp_max > 2.2:
-        print("%s-%s-%s-filtered" % (idx0, idx1, tmp_max))
+        print("%s-filtered" % (wav_name, tmp_max))
         return
     tmp_audio32 = (tmp_audio / tmp_max * (maxx * alpha*32768)) + ((1 - alpha)*32768) * tmp_audio
     tmp_audio = librosa.resample(
         tmp_audio32, orig_sr=32000, target_sr=16000
-    )
+    )#不是重采样问题
     tensor_wav16 = torch.from_numpy(tmp_audio)
     if (is_half == True):
         tensor_wav16=tensor_wav16.half().to(device)
     else:
         tensor_wav16 = tensor_wav16.to(device)
     ssl=model.model(tensor_wav16.unsqueeze(0))["last_hidden_state"].transpose(1,2).cpu()#torch.Size([1, 768, 215])
-    if np.isnan(ssl.detach().numpy()).sum()!= 0:return
+    if np.isnan(ssl.detach().numpy()).sum()!= 0:
+        nan_fails.append(wav_name)
+        print("nan filtered:%s"%wav_name)
+        return
     wavfile.write(
         "%s/%s"%(wav32dir,wav_name),
         32000,
         tmp_audio32.astype("int16"),
     )
-    # torch.save(ssl,hubert_path )
     my_save(ssl,hubert_path )
 
 with open(inp_text,"r",encoding="utf8")as f:
@@ -92,3 +97,12 @@ for line in lines[int(i_part)::int(all_parts)]:
         name2go(wav_name)
     except:
         print(line,traceback.format_exc())
+
+if(len(nan_fails)>0 and is_half==True):
+    is_half=False
+    model=model.float()
+    for wav_name in nan_fails:
+        try:
+            name2go(wav_name)
+        except:
+            print(wav_name,traceback.format_exc())

docs/cn/Changelog_CN.md

@@ -20,5 +20,6 @@
 
 3-音频路径检查。如果尝试读取输入错的路径报错路径不存在，而非ffmpeg错误。
 
-待修复：-hubert提取在half下出现nan概率更高的问题
+### 20240123更新
+1-hubert提取在half下出现nan概率更高的问题