Update app.py

app.py

@@ -48,9 +48,11 @@ def transcribe_audio(audio_file):
 
     start_time = time.time()
 
-    # Load the audio file and process it with Whisper's processor
-    audio, sample_rate = whisper_processor.audio_to_array(audio_file)
-    input_features = whisper_processor(audio, sampling_rate=sample_rate, return_tensors="pt").input_features.to(device)
+    # Load the audio file using torchaudio
+    waveform, sample_rate = torchaudio.load(audio_file)
+    
+    # Process the waveform with Whisper's processor
+    input_features = whisper_processor(waveform, sampling_rate=sample_rate, return_tensors="pt").input_features.to(device)
 
     # Generate the transcription
     output = whisper_model.generate(input_features=input_features)
@@ -62,6 +64,7 @@ def transcribe_audio(audio_file):
     result = f"Time taken: {output_time:.2f} seconds\nNumber of words: {len(text.split())}"
 
     return text, result
+
 # Clean and preprocess text for summarization
 def clean_text(text):
     text = re.sub(r'https?:\/\/.*[\r\n]*', '', text)