Remove warnings and add some warnings (#1408)

GPT_SoVITS/feature_extractor/cnhubert.py

@@ -4,6 +4,7 @@ import librosa
 import torch
 import torch.nn.functional as F
 import soundfile as sf
+import os
 from transformers import logging as tf_logging
 tf_logging.set_verbosity_error()
 
@@ -24,9 +25,11 @@ cnhubert_base_path = None
 class CNHubert(nn.Module):
     def __init__(self):
         super().__init__()
-        self.model = HubertModel.from_pretrained(cnhubert_base_path)
+        if os.path.exists(cnhubert_base_path):...
+        else:raise FileNotFoundError(cnhubert_base_path)
+        self.model = HubertModel.from_pretrained(cnhubert_base_path, local_files_only=True)
         self.feature_extractor = Wav2Vec2FeatureExtractor.from_pretrained(
-            cnhubert_base_path
+            cnhubert_base_path, local_files_only=True
         )
 
     def forward(self, x):

GPT_SoVITS/inference_webui.py

@@ -389,6 +389,10 @@ def merge_short_text_in_array(texts, threshold):
 cache= {}
 def get_tts_wav(ref_wav_path, prompt_text, prompt_language, text, text_language, how_to_cut=i18n("不切"), top_k=20, top_p=0.6, temperature=0.6, ref_free = False,speed=1,if_freeze=False):
     global cache
+    if ref_wav_path:pass
+    else:gr.Warning(i18n('请上传参考音频'))
+    if text:pass
+    else:gr.Warning(i18n('请填入推理文本'))
     t = []
     if prompt_text is None or len(prompt_text) == 0:
         ref_free = True
@@ -413,6 +417,7 @@ def get_tts_wav(ref_wav_path, prompt_text, prompt_language, text, text_language,
         with torch.no_grad():
             wav16k, sr = librosa.load(ref_wav_path, sr=16000)
             if (wav16k.shape[0] > 160000 or wav16k.shape[0] < 48000):
+                gr.Warning(i18n("参考音频在3~10秒范围外，请更换！"))
                 raise OSError(i18n("参考音频在3~10秒范围外，请更换！"))
             wav16k = torch.from_numpy(wav16k)
             zero_wav_torch = torch.from_numpy(zero_wav)

GPT_SoVITS/prepare_datasets/1-get-text.py

@@ -54,6 +54,8 @@ if os.path.exists(txt_path) == False:
     #     device = "mps"
     else:
         device = "cpu"
+    if os.path.exists(bert_pretrained_dir):...
+    else:raise FileNotFoundError(bert_pretrained_dir)
     tokenizer = AutoTokenizer.from_pretrained(bert_pretrained_dir)
     bert_model = AutoModelForMaskedLM.from_pretrained(bert_pretrained_dir)
     if is_half == True:

GPT_SoVITS/prepare_datasets/3-get-semantic.py

@@ -34,6 +34,8 @@ logging.getLogger("numba").setLevel(logging.WARNING)
 # os.environ["CUDA_VISIBLE_DEVICES"]=sys.argv[5]
 # opt_dir="/data/docker/liujing04/gpt-vits/fine_tune_dataset/%s"%exp_name
 
+if os.path.exists(pretrained_s2G):...
+else:raise FileNotFoundError(pretrained_s2G)
 
 hubert_dir = "%s/4-cnhubert" % (opt_dir)
 semantic_path = "%s/6-name2semantic-%s.tsv" % (opt_dir, i_part)