diff --git a/tools/asr/fasterwhisper_asr.py b/tools/asr/fasterwhisper_asr.py
index 669ac3aa..d749a798 100644
--- a/tools/asr/fasterwhisper_asr.py
+++ b/tools/asr/fasterwhisper_asr.py
@@ -34,6 +34,19 @@ 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]
@@ -57,6 +70,8 @@ def execute_asr(input_folder, output_folder, model_size, language, precision):
     
     for file_name in tqdm(input_file_names):
         try:
+            if file_name.split(".")[-1] not in AUDIO_EXTENSIONS:
+                continue
             file_path = os.path.join(input_folder, file_name)
             segments, info = model.transcribe(
                 audio          = file_path,
diff --git a/tools/asr/funasr_asr.py b/tools/asr/funasr_asr.py
index 831da6c1..40bfe2e2 100644
--- a/tools/asr/funasr_asr.py
+++ b/tools/asr/funasr_asr.py
@@ -23,6 +23,19 @@ 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"]
@@ -40,6 +53,8 @@ def execute_asr(input_folder, output_folder, model_size, language):
 
     for file_name in tqdm(input_file_names):
         try:
+            if file_name.split(".")[-1] not in AUDIO_EXTENSIONS:
+                continue
             file_path = os.path.join(input_folder, file_name)
             text = model.generate(input=file_path)[0]["text"]
             output.append(f"{file_path}|{output_file_name}|{language.upper()}|{text}")
diff --git a/tools/cmd-denoise.py b/tools/cmd-denoise.py
index 69b51e66..457cf6a9 100644
--- a/tools/cmd-denoise.py
+++ b/tools/cmd-denoise.py
@@ -4,6 +4,19 @@ 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",
+]
+
 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"
 ans = pipeline(Tasks.acoustic_noise_suppression,model=path_denoise)
@@ -12,6 +25,8 @@ def execute_denoise(input_folder,output_folder):
     # print(input_folder)
     # print(list(os.listdir(input_folder).sort()))
     for name in tqdm(os.listdir(input_folder)):
+        if name.split(".")[-1] not in AUDIO_EXTENSIONS:
+            continue
         ans("%s/%s"%(input_folder,name),output_path='%s/%s'%(output_folder,name))
 
 if __name__ == '__main__':
diff --git a/tools/slice_audio.py b/tools/slice_audio.py
index 46ee408a..4969d2cb 100644
--- a/tools/slice_audio.py
+++ b/tools/slice_audio.py
@@ -6,6 +6,19 @@ from scipy.io import wavfile
 from my_utils import load_audio
 from slicer2 import Slicer
 
+AUDIO_EXTENSIONS = [
+    "mp3",
+    "wav",
+    "flac",
+    "ogg",
+    "m4a",
+    "wma",
+    "aac",
+    "aiff",
+    "aif",
+    "aifc",
+]
+
 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)
     if os.path.isfile(inp):
@@ -28,6 +41,8 @@ def slice(inp,opt_root,threshold,min_length,min_interval,hop_size,max_sil_kept,_
         # print(inp_path)
         try:
             name = os.path.basename(inp_path)
+            if name.split(".")[-1] not in AUDIO_EXTENSIONS:
+                continue
             audio = load_audio(inp_path, 32000)
             # print(audio.shape)
             for chunk, start, end in slicer.slice(audio):  # start和end是帧数
diff --git a/tools/uvr5/webui.py b/tools/uvr5/webui.py
index a690a686..c403c89f 100644
--- a/tools/uvr5/webui.py
+++ b/tools/uvr5/webui.py
@@ -23,15 +23,28 @@ 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:
-        inp_root = inp_root.strip(" ").strip('"').strip("\n").strip('"').strip(" ")
+        inp_root = inp_root.strip(" ").strip('"').strip("\n").strip('"').strip(" ").strip("\u202a")
         save_root_vocal = (
-            save_root_vocal.strip(" ").strip('"').strip("\n").strip('"').strip(" ")
+            save_root_vocal.strip(" ").strip('"').strip("\n").strip('"').strip(" ").strip("\u202a")
         )
         save_root_ins = (
-            save_root_ins.strip(" ").strip('"').strip("\n").strip('"').strip(" ")
+            save_root_ins.strip(" ").strip('"').strip("\n").strip('"').strip(" ").strip("\u202a")
         )
         is_hp3 = "HP3" in model_name
         if model_name == "onnx_dereverb_By_FoxJoy":
@@ -49,6 +62,8 @@ def uvr(model_name, inp_root, save_root_vocal, paths, save_root_ins, agg, format
         else:
             paths = [path.name for path in paths]
         for path in paths:
+            if path.split(".")[-1] not in AUDIO_EXTENSIONS:
+                continue
             inp_path = os.path.join(inp_root, path)
             if(os.path.isfile(inp_path)==False):continue
             need_reformat = 1