Update app.py

app.py

@@ -1,7 +1,7 @@
 import gradio as gr
 from transformers import BlipProcessor, BlipForConditionalGeneration
-from gtts import gTTS
-import IPython.display as ipd
+from pydub import AudioSegment
+from pydub.playback import play
 
 model_id = "dblasko/blip-dalle3-img2prompt"
 model = BlipForConditionalGeneration.from_pretrained(model_id)
@@ -21,9 +21,10 @@ def generate_caption(image):
 
     return generated_caption, audio_path
 
+
 def play_audio(audio_path):
-    # Display an audio player
-    return ipd.Audio(audio_path)
+    sound = AudioSegment.from_file(audio_path)
+    play(sound)
 
 # Create a Gradio interface with an image input, a textbox output, a button, and an audio player
 demo = gr.Interface(