Update app.py

app.py

@@ -43,9 +43,21 @@ The model was trained on Thai audio recordings with the following sentences: \n
 
 
 
-def predict(file_path):
+def predict(file_upload,microphone_path):
     max_length = 100000
+    file_path = file_upload
+    if (microphone is not None) and (file_upload is not None):
+        warn_output = (
+            "WARNING: You've uploaded an audio file and used the microphone. "
+            "The recorded file from the microphone will be used and the uploaded audio will be discarded.\n"
+        )
 
+    elif (microphone is None) and (file_upload is None):
+        return "ERROR: You have to either use the microphone or upload an audio file"
+    if(microphone is not None):
+        file_path = microphone
+    if(file_upload is not None):
+        file_path = file_upload
     model.eval()
     with torch.no_grad():
         wav_data, _ = sf.read(file_path.name)
@@ -64,9 +76,13 @@ def predict(file_path):
         predicted_class_id = torch.argmax(logits, dim=-1).item()
 
     return predicted_class_id
+    
 gr.Interface(
     fn=predict,
-    inputs="file",
+    inputs=[
+        gr.inputs.Audio(source="upload", type="filepath", optional=True),
+        gr.inputs.Audio(source="microphone", type="filepath", optional=True),
+    ],
     outputs="text",
     title=title,
     description=description,