Update app.py

app.py

@@ -4,6 +4,8 @@ import streamlit as st
 from transformers import pipeline
 from PIL import Image
 import numpy as np
+import time
+import threading
 
 # ——— 1) MODEL LOADING (cached) ————————————————
 @st.cache_resource
@@ -104,7 +106,7 @@ st.markdown("""
 }
 .stButton>button {
     background-color: #ffcccb;
-    color: #000000;
+    button-color: #000000;
     border-radius: 10px;
     border: 2px solid #ff9999;
     font-size: 18px;
@@ -198,12 +200,18 @@ with st.container():
     st.markdown("<div class='section-header'>2️⃣ What's in the Picture? 🧐</div>", unsafe_allow_html=True)
     captioner = get_image_captioner()
     progress_bar = st.progress(0)
+    result = [None]
+    def run_caption():
+        result[0] = part1_image_to_text(pil_img, captioner)
     with st.spinner("Figuring out what's in your picture..."):
+        thread = threading.Thread(target=run_caption)
+        thread.start()
         for i in range(100):
             progress_bar.progress(i + 1)
-            if i == 99:
-                caption = part1_image_to_text(pil_img, captioner)
+            time.sleep(0.05)  # Adjust for ~5 seconds total
+        thread.join()
     progress_bar.empty()
+    caption = result[0]
     st.markdown(f"<div class='caption-box'><b>Picture Description:</b><br>{caption}</div>", unsafe_allow_html=True)
 
 # Story and audio section
@@ -212,21 +220,33 @@ with st.container():
     # Story
     story_pipe = get_story_pipe()
     progress_bar = st.progress(0)
+    result = [None]
+    def run_story():
+        result[0] = part2_text_to_story(caption, story_pipe)
     with st.spinner("Writing a super cool story..."):
+        thread = threading.Thread(target=run_story)
+        thread.start()
         for i in range(100):
             progress_bar.progress(i + 1)
-            if i == 99:
-                story = part2_text_to_story(caption, story_pipe)
+            time.sleep(0.07)  # Adjust for ~7 seconds total
+        thread.join()
     progress_bar.empty()
+    story = result[0]
     st.markdown(f"<div class='story-box'><b>Your Cool Story! 📚</b><br>{story}</div>", unsafe_allow_html=True)
 
     # TTS
     tts_pipe = get_tts_pipe()
     progress_bar = st.progress(0)
+    result = [None]
+    def run_tts():
+        result[0] = part3_text_to_speech_bytes(story, tts_pipe)
     with st.spinner("Turning your story into sound..."):
+        thread = threading.Thread(target=run_tts)
+        thread.start()
         for i in range(100):
             progress_bar.progress(i + 1)
-            if i == 99:
-                audio_bytes = part3_text_to_speech_bytes(story, tts_pipe)
+            time.sleep(0.10)  # Adjust for ~10 seconds total
+        thread.join()
     progress_bar.empty()
+    audio_bytes = result[0]
     st.audio(audio_bytes, format="audio/wav")