Upadte config.py

config.py

@@ -15,6 +15,19 @@ bert_path = "GPT_SoVITS/pretrained_models/chinese-roberta-wwm-ext-large"
 pretrained_sovits_path = "GPT_SoVITS/pretrained_models/s2G488k.pth"
 pretrained_gpt_path = "GPT_SoVITS/pretrained_models/s1bert25hz-2kh-longer-epoch=68e-step=50232.ckpt"
 
+AUDIO_EXTENSIONS = [
+    "mp3",
+    "wav",
+    "flac",
+    "ogg",
+    "m4a",
+    "wma",
+    "aac",
+    "aiff",
+    "aif",
+    "aifc",
+]
+
 exp_root = "logs"
 python_exec = sys.executable or "python"
 if torch.cuda.is_available():

tools/asr/fasterwhisper_asr.py

@@ -10,6 +10,7 @@ from faster_whisper import WhisperModel
 from tqdm import tqdm
 
 from tools.asr.config import check_fw_local_models
+from config import AUDIO_EXTENSIONS
 
 language_code_list = [
     "af", "am", "ar", "as", "az", 
@@ -34,19 +35,6 @@ language_code_list = [
     "vi", "yi", "yo", "zh", "yue",
     "auto"]
 
-AUDIO_EXTENSIONS = [
-    "mp3",
-    "wav",
-    "flac",
-    "ogg",
-    "m4a",
-    "wma",
-    "aac",
-    "aiff",
-    "aif",
-    "aifc",
-]
-
 def execute_asr(input_folder, output_folder, model_size, language, precision):
     if '-local' in model_size:
         model_size = model_size[:-6]

tools/asr/funasr_asr.py

@@ -4,6 +4,7 @@ import argparse
 import os
 import traceback
 from tqdm import tqdm
+from config import AUDIO_EXTENSIONS
 
 from funasr import AutoModel
 
@@ -23,19 +24,6 @@ model = AutoModel(
     punc_model_revision = "v2.0.4",
 )
 
-AUDIO_EXTENSIONS = [
-    "mp3",
-    "wav",
-    "flac",
-    "ogg",
-    "m4a",
-    "wma",
-    "aac",
-    "aiff",
-    "aif",
-    "aifc",
-]
-
 def only_asr(input_file):
     try:
         text = model.generate(input=input_file)[0]["text"]

tools/cmd-denoise.py

@@ -3,19 +3,7 @@ import os,argparse
 from modelscope.pipelines import pipeline
 from modelscope.utils.constant import Tasks
 from tqdm import tqdm
-
-AUDIO_EXTENSIONS = [
-    "mp3",
-    "wav",
-    "flac",
-    "ogg",
-    "m4a",
-    "wma",
-    "aac",
-    "aiff",
-    "aif",
-    "aifc",
-]
+from config import AUDIO_EXTENSIONS
 
 path_denoise  = 'tools/denoise-model/speech_frcrn_ans_cirm_16k'
 path_denoise  = path_denoise  if os.path.exists(path_denoise)  else "damo/speech_frcrn_ans_cirm_16k"

tools/slice_audio.py

@@ -5,19 +5,7 @@ from scipy.io import wavfile
 # sys.path.append(parent_directory)
 from my_utils import load_audio
 from slicer2 import Slicer
-
-AUDIO_EXTENSIONS = [
-    "mp3",
-    "wav",
-    "flac",
-    "ogg",
-    "m4a",
-    "wma",
-    "aac",
-    "aiff",
-    "aif",
-    "aifc",
-]
+from config import AUDIO_EXTENSIONS
 
 def slice(inp,opt_root,threshold,min_length,min_interval,hop_size,max_sil_kept,_max,alpha,i_part,all_part):
     os.makedirs(opt_root,exist_ok=True)

tools/uvr5/webui.py

@@ -2,6 +2,7 @@ import os
 import traceback,gradio as gr
 import logging
 from tools.i18n.i18n import I18nAuto
+from config import AUDIO_EXTENSIONS
 i18n = I18nAuto()
 
 logger = logging.getLogger(__name__)
@@ -23,19 +24,6 @@ is_half=eval(sys.argv[2])
 webui_port_uvr5=int(sys.argv[3])
 is_share=eval(sys.argv[4])
 
-AUDIO_EXTENSIONS = [
-    "mp3",
-    "wav",
-    "flac",
-    "ogg",
-    "m4a",
-    "wma",
-    "aac",
-    "aiff",
-    "aif",
-    "aifc",
-]
-
 def uvr(model_name, inp_root, save_root_vocal, paths, save_root_ins, agg, format0):
     infos = []
     try: