Update app.py

app.py

@@ -12,11 +12,13 @@ processor = WhisperProcessor.from_pretrained("openai/whisper-large")
 model = WhisperForConditionalGeneration.from_pretrained("openai/whisper-large")
 model.config.forced_decoder_ids = None
 
-# Function to perform ASR on audio data
-def transcribe_audio(audio_data):
-    # Process audio data using the Whisper processor
-    input_features = processor(audio_data, return_tensors="pt").input_features 
+# 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):
     # Generate token ids
     predicted_ids = model.generate(input_features)
     
@@ -25,11 +27,6 @@ def transcribe_audio(audio_data):
     
     return transcription[0]
 
-# Custom preprocessing function
-def preprocess_audio(audio_data):
-    # Apply any custom preprocessing to the audio data here if needed
-    return audio_data
-
 # Create Gradio interface
 audio_input = gr.Audio(preprocess=preprocess_audio)
-gr.Interface(fn=transcribe_audio, inputs=audio_input, outputs="text").launch()
+gr.Interface(fn=transcribe_audio, inputs=audio_input, outputs="text").launch()