Update app.py

app.py

@@ -1,7 +1,7 @@
 import gradio as gr
 from transformers import BlipProcessor, BlipForConditionalGeneration
 from gtts import gTTS
-import pygame
+from IPython.display import Audio
 
 model_id = "dblasko/blip-dalle3-img2prompt"
 model = BlipForConditionalGeneration.from_pretrained(model_id)
@@ -22,10 +22,7 @@ def generate_caption(image):
     return generated_caption, audio_path
 
 def play_audio(audio_path):
-    pygame.mixer.init()
-    pygame.mixer.music.load(audio_path)
-    pygame.mixer.music.play()
-    pygame.event.wait()
+    return Audio(audio_path)
 
 # Create a Gradio interface with an image input, a textbox output, a button, and an audio player
 demo = gr.Interface(
@@ -34,7 +31,6 @@ demo = gr.Interface(
     outputs=[
         gr.Textbox(label="Generated caption"), 
         gr.Button("Convert to Audio", play_audio),
-        gr.Audio("audio")
     ]
 )
 demo.launch()