Update app.py

app.py

@@ -2,12 +2,8 @@ import gradio as gr
 from transformers import BlipProcessor, BlipForConditionalGeneration
 from gtts import gTTS
 from playsound import playsound
-from transformers import pipeline
+from concurrent.futures import ThreadPoolExecutor
 
-# Load the text-to-speech model
-tts_synthesizer = pipeline("text-to-speech", "suno/bark")
-
-# Load the image-to-text model
 model_id = "dblasko/blip-dalle3-img2prompt"
 model = BlipForConditionalGeneration.from_pretrained(model_id)
 processor = BlipProcessor.from_pretrained(model_id)
@@ -20,10 +16,9 @@ def generate_caption(image):
     generated_caption = processor.batch_decode(generated_ids, skip_special_tokens=True, temperature=0.8, top_k=40, top_p=0.9)[0]
 
     # Convert the generated caption to speech
-    speech = tts_synthesizer(generated_caption)
-    audio_path = "generated_audio.wav"
-    with open(audio_path, "wb") as f:
-        f.write(speech["audio"])
+    tts = gTTS(text=generated_caption, lang='en')
+    audio_path = "generated_audio.mp3"
+    tts.save(audio_path)
 
     return generated_caption, audio_path
 
@@ -37,6 +32,7 @@ demo = gr.Interface(
     outputs=[
         gr.Textbox(label="Generated caption"), 
         gr.Button("Convert to Audio", play_audio),
-    ]
+    ],
+    live=True  # ทำให้ Gradio ทำงานแบบไม่บล็อก
 )
 demo.launch(share=True)