Update app.py

app.py

@@ -12,14 +12,16 @@ processor = WhisperProcessor.from_pretrained("openai/whisper-large")
 model = WhisperForConditionalGeneration.from_pretrained("openai/whisper-large")
 forced_decoder_ids = processor.get_decoder_prompt_ids(language="italian", task="transcribe")
 
-
 # Custom preprocessing function
 def preprocess_audio(audio_data):
     # Apply any custom preprocessing to the audio data here if needed
     return processor(audio_data, return_tensors="pt").input_features
 
 # Function to perform ASR on audio data
-def transcribe_audio(input_features):
+def transcribe_audio(audio_data):
+    # Preprocess the audio data
+    input_features = preprocess_audio(audio_data)
+
     # Generate token ids
     predicted_ids = model.generate(input_features)
     
@@ -29,5 +31,5 @@ def transcribe_audio(input_features):
     return transcription[0]
 
 # Create Gradio interface
-audio_input = gr.Audio(preprocess=preprocess_audio)
-gr.Interface(fn=transcribe_audio, inputs=audio_input, outputs="text").launch()
+audio_input = gr.Audio()
+gr.Interface(fn=transcribe_audio, inputs=audio_input, outputs="text").launch()