Update app.py

app.py

@@ -11,23 +11,7 @@ import matplotlib.pyplot as plt
 import gradio as gr
 
 
-
-def inference(audio)
-  speech, rate = soundfile.read(audio.name)
-  assert rate == fs, "mismatch in sampling rate"
-  nbests = speech2text(speech)
-  text, *_ = nbests[0]
-
-  print(f"Input Speech: {file_name}")
-  display(Audio(speech, rate=rate))
-  librosa.display.waveplot(speech, sr=rate)
-  plt.show()
-  print(f"ASR hypothesis: {text_normalizer(text)}")
-  print("*" * 50)
-
-
 d = ModelDownloader()
-# It may takes a while to download and build models
 speech2text = Speech2Text(
     **d.download_and_unpack(tag),
     device="cpu",
@@ -45,4 +29,20 @@ def text_normalizer(text):
     
 lang = 'multilingual'
 fs = 16000 
-tag = 'ftshijt/open_li52_asr_train_asr_raw_bpe7000_valid.acc.ave_10best' 
+tag = 'ftshijt/open_li52_asr_train_asr_raw_bpe7000_valid.acc.ave_10best' 
+
+
+def inference(audio):
+  speech, rate = soundfile.read(audio.name)
+  assert rate == fs, "mismatch in sampling rate"
+  nbests = speech2text(speech)
+  text, *_ = nbests[0]
+
+  print(f"Input Speech: {file_name}")
+  display(Audio(speech, rate=rate))
+  librosa.display.waveplot(speech, sr=rate)
+  plt.show()
+  print(f"ASR hypothesis: {text_normalizer(text)}")
+  print("*" * 50)
+
+