Update app.py

app.py

@@ -10,10 +10,7 @@ processor = Wav2Vec2Processor.from_pretrained("aditii09/facebook_english_asr")
 model = Wav2Vec2ForCTC.from_pretrained("aditii09/facebook_english_asr")
 
 def speech2text(audio):
-    if isinstance(audio, tuple) and len(audio) == 2:
-        sr, data = audio
-    else:
-        raise ValueError("Invalid audio format. Expected (sr, data) tuple.")
+    sr, data = audio
 
     # resample to 16hz
     data_16hz = librosa.resample(data[:,0].astype(np.float32),sr,16000)
@@ -30,11 +27,6 @@ def speech2text(audio):
 
     return transcription[0].lower()  # batch size 1
 
-iface = gr.Interface(
-    fn=speech2text,
-    inputs=gr.Audio(source="upload", type="filepath"),
-    examples=[["example.flac"]],
-    outputs="text"
-)
+iface = gr.Interface(speech2text, "upload", "text")
 
-iface.launch()
+iface.launch()