Update app.py

app.py

@@ -22,8 +22,14 @@ def process_audio_file(file):
     return inputs
 
 
-def interpret(file):
-    inputs = process_audio_file(file)
+def interpret(micro, file):
+    if file is not None and micro is None:
+        input_audio = file
+    elif file is None and micro is not None:
+        input_audio = micro
+    else:
+        input_audio = file
+    inputs = process_audio_file(input_audio)
     with torch.no_grad():
         output_logit = model(inputs.input_values).logits
     pred_ids = torch.argmax(output_logit, dim=-1)
@@ -46,10 +52,11 @@ iface = gr.Interface(
     description="A simple interface to interpret from spoken English to Vietnamese.",
     article="Author: <a href=\"https://huggingface.co/vumichien\">Vu Minh Chien</a>.", 
     inputs=[
-        gr.Audio(source="microphone", type="filepath", streaming=False), 
-        gr.Audio(source="upload", type="filepath", optional=True),
+        gr.Audio(source="microphone", type="filepath", streaming=False, label="Microphone"), 
+        gr.Audio(source="upload", type="filepath", label="File"),
     ],
-    outputs=gr.outputs.Textbox(label="Interpreted text")
+    outputs=gr.Textbox(label="Text output")
     ,
+    samples=["samples/sample-000000.mp3", "samples/sample-000001.mp3"]
     )
-iface.launch(enable_queue=True, debug=True)
+iface.launch(debug=True)