update app.py

app.py

@@ -189,8 +189,8 @@ audio_output_prediction, sample_rate, min_duration, frame_length, hop_length_fra
     print(audio_output_prediction)
     sf.write(audio_output_prediction , denoise_long[0, :], sample_rate)
 
-def denoise_audio(audioName,mic):
-  if audioName != 0:
+def denoise_audio(audioName):
+  
     sr, data = audioName
     sf.write("temp.wav",data, sr)
     testNo = "temp"
@@ -200,36 +200,29 @@ def denoise_audio(audioName,mic):
 
     
     
-  else:
-    sr, data = mic
-    sf.write("temp.wav",data, sr)
-    testNo = "temp"
-    audio_dir_prediction = os.path.abspath("/")+ str(testNo) +".wav"
-    sample_rate, data = mic[0], mic[1]
-    len_data = len(data)
   
 
-  t = len_data / sample_rate # returns duration but in floats
-  print("t:",t)
-  weights_path = os.path.abspath("./")
-  name_model = "model_unet"
-  audio_dir_prediction = os.path.abspath("./")
-  dir_save_prediction = os.path.abspath("./")
-  audio_output_prediction = "test.wav"
-  audio_input_prediction = ["temp.wav"]
-  sample_rate = 8000
-  min_duration = t
-  frame_length = 8064
-  hop_length_frame = 8064
-  n_fft = 255
-  hop_length_fft = 63
-
-  dim_square_spec = int(n_fft / 2) + 1
-
-  prediction(weights_path, name_model, audio_dir_prediction, dir_save_prediction, audio_input_prediction,
-            audio_output_prediction, sample_rate, min_duration, frame_length, hop_length_frame, n_fft, hop_length_fft)
-  print(audio_output_prediction)
-  return audio_output_prediction
+    t = len_data / sample_rate # returns duration but in floats
+    print("t:",t)
+    weights_path = os.path.abspath("./")
+    name_model = "model_unet"
+    audio_dir_prediction = os.path.abspath("./")
+    dir_save_prediction = os.path.abspath("./")
+    audio_output_prediction = "test.wav"
+    audio_input_prediction = ["temp.wav"]
+    sample_rate = 8000
+    min_duration = t
+    frame_length = 8064
+    hop_length_frame = 8064
+    n_fft = 255
+    hop_length_fft = 63
+
+    dim_square_spec = int(n_fft / 2) + 1
+
+    prediction(weights_path, name_model, audio_dir_prediction, dir_save_prediction, audio_input_prediction,
+                audio_output_prediction, sample_rate, min_duration, frame_length, hop_length_frame, n_fft, hop_length_fft)
+    print(audio_output_prediction)
+    return audio_output_prediction
 
 
 examples = [
@@ -241,7 +234,7 @@ examples = [
 
 
 iface = gr.Interface(fn = denoise_audio,
-                     inputs = ['audio',gr.inputs.Audio(source="microphone", type="filepath", label="Record your audio")],
+                     inputs = 'audio',
                      outputs = 'audio',
                      title = 'audio to denoised Audio Application',
                      description = 'A simple application to denoise audio speech usinf UNet deep learning model. Upload your own audio, or click one of the examples to load them.',