Update app.py

app.py

@@ -6,39 +6,35 @@ import soundfile as sf
 import tempfile
 import os
 from PIL import Image
+import string
 
-# Initialize pipelines with caching to avoid reloading
+# Initialize pipelines with caching
 @st.cache_resource
 def load_pipelines():
-    # Load pipeline for generating captions from images
     captioner = pipeline("image-to-text", model="Salesforce/blip-image-captioning-large")
-    # Load pipeline for generating stories from text prompts
     storyer = pipeline("text-generation", model="aspis/gpt2-genre-story-generation")
-    # Load pipeline for converting text to speech
     tts = pipeline("text-to-speech", model="facebook/mms-tts-eng")
     return captioner, storyer, tts
 
-# Load the pipelines once and reuse them
 captioner, storyer, tts = load_pipelines()
 
-# Function to generate caption, story, and audio from an uploaded image
+# Function to generate content from an image
 def generate_content(image):
-    # Convert the uploaded image to a PIL image format
     pil_image = Image.open(image)
     
-    # Generate a caption based on the image content
+    # Generate caption
     caption = captioner(pil_image)[0]["generated_text"]
     st.write("**🌟 What's in the picture: 🌟**")
     st.write(caption)
 
-    # Create a prompt for generating a children's story
+    # Create prompt for story
     prompt = (
         f"Write a funny, warm children's story for ages 3-10, 50–100 words, "
         f"in third-person narrative, that describes this scene exactly: {caption} "
         f"mention the exact place, location or venue within {caption}"
     )
     
-    # Generate the story based on the prompt
+    # Generate raw story
     raw = storyer(
         prompt,
         max_new_tokens=150,
@@ -48,52 +44,42 @@ def generate_content(image):
         return_full_text=False
     )[0]["generated_text"].strip()
 
-    # Trim the generated story to a maximum of 100 words
-    words = raw.split()
+    # Define allowed characters to keep (removes symbols like * and ~)
+    allowed_chars = string.ascii_letters + string.digits + " .,!?\"'-"
+    
+    # Clean the raw story by keeping only allowed characters
+    clean_raw = ''.join(c for c in raw if c in allowed_chars)
+    
+    # Split into words and trim to 100 words
+    words = clean_raw.split()
     story = " ".join(words[:100])
+    
     st.write("**📖 Your funny story: 📖**")
     st.write(story)
 
-    # Split the story into chunks of 200 characters for text-to-speech processing
+    # Generate audio from cleaned story
     chunks = textwrap.wrap(story, width=200)
-    
-    # Generate and concatenate audio for each text chunk
     audio = np.concatenate([tts(chunk)["audio"].squeeze() for chunk in chunks])
 
-    # Save the concatenated audio to a temporary WAV file
+    # Save audio to temporary file
     with tempfile.NamedTemporaryFile(delete=False, suffix=".wav") as temp_file:
         sf.write(temp_file.name, audio, tts.model.config.sampling_rate)
         temp_file_path = temp_file.name
 
     return caption, story, temp_file_path
 
-# Streamlit UI for the application
-st.markdown(
-    """
-    <style>
-    .stApp {
-        background: linear-gradient(to right, #e0f7fa, #b2ebf2);
-    }
-    </style>
-    """,
-    unsafe_allow_html=True
-)
-
+# Streamlit UI
 st.title("✨ Magic Story Maker ✨")
 st.markdown("Upload a picture to make a funny story and hear it too! 📸")
 
-# File uploader for image input
-uploaded_image = st.file_uploader("Choose your picture", type=["jpg", "jpeg", "png"], help="Pick a photo to start the magic!")
-
-# Placeholder image URL (replace with an actual URL of a child-friendly image)
-placeholder_url = "https://example.com/placeholder_image.jpg"
+uploaded_image = st.file_uploader("Choose your picture", type=["jpg", "jpeg", "png"])
 
 if uploaded_image is None:
-    st.image(placeholder_url, caption="Upload your picture here! 📷", use_column_width=True)
+    st.image("https://example.com/placeholder_image.jpg", caption="Upload your picture here! 📷", use_column_width=True)
 else:
     st.image(uploaded_image, caption="Your Picture 🌟", use_column_width=True)
 
-if st.button("✨ Make My Story! ✨", help="Click to create your magic story"):
+if st.button("✨ Make My Story! ✨"):
     if uploaded_image is not None:
         with st.spinner("🔮 Creating your magical story..."):
             caption, story, audio_path = generate_content(uploaded_image)