Update app.py

app.py

@@ -34,13 +34,15 @@ def handle_image(img):
 
 # Function to handle audio input
 def handle_audio(audio):
-    speech, _ = sf.read(audio)
-    input_values = processor(speech, return_tensors="pt").input_values
+    # Gradio's Audio component returns a tuple of (sample_rate, audio_data)
+    sample_rate, audio_data = audio
+    input_values = wav2vec2_processor(audio_data, sampling_rate=sample_rate, return_tensors="pt").input_values
     logits = wav2vec2_model(input_values).logits
     predicted_ids = torch.argmax(logits, dim=-1)
-    transcriptions = processor.decode(predicted_ids[0])
+    transcriptions = wav2vec2_processor.decode(predicted_ids[0])
     return handle_text(transcriptions)
 
+
 def chatbot(text, img, audio):
     text_output = handle_text(text) if text is not None else ''
     img_output = handle_image(img) if img is not None else ''