Update app.py

app.py

@@ -16,17 +16,28 @@ def process_audio_file(file):
     input_values = feature_extractor(data, return_tensors="pt").input_values
     return input_values
     
-def transcribe(file, target_language):
+def transcribe(file_mic, file_upload, target_language):
     
     target_code = target_language.split("(")[-1].split(")")[0]
     forced_bos_token_id = MAPPING[target_code]
-
+    
+    warn_output = ""
+    if (file_mic 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"
+       file = file_mic
+    elif (file_mic is None) and (file_upload is None):
+       return "ERROR: You have to either use the microphone or upload an audio file"
+    elif file_mic is not None:
+       file = file_mic
+    else:
+       file = file_upload
+       
     input_values = process_audio_file(file)
     
-    sequences = model.generate(input_values, forced_bos_token_id=forced_bos_token_id)
+    sequences = model.generate(input_values, forced_bos_token_id=forced_bos_token_id, num_beams=1, max_length=30)
     
     transcription = tokenizer.batch_decode(sequences, skip_special_tokens=True)
-    return transcription[0]
+    return warn_output + transcription[0]
     
 target_language = [
     "English (en)",
@@ -69,7 +80,8 @@ MAPPING = {
 iface = gr.Interface(
     fn=transcribe, 
     inputs=[
-        gr.inputs.Audio(source="microphone", type='filepath'),
+        gr.inputs.Audio(source="microphone", type='filepath', optional=True),
+        gr.inputs.Audio(source="upload", type='filepath', optional=True),
         gr.inputs.Dropdown(target_language),
     ],
     outputs="text",