Update app.py

app.py

@@ -1,5 +1,6 @@
 import streamlit as st
 from transformers import pipeline
+import hashlib
 
 # Function definitions
 def img2text(url):
@@ -28,55 +29,50 @@ def main():
     st.set_page_config(page_title="Your Image to Audio Story", page_icon="🦜")
     st.header("Turn Your Image to Audio Story")
     
-    # File uploader
     uploaded_file = st.file_uploader("Select an Image...")
     
     if uploaded_file is not None:
-        # Save the uploaded file locally.
+        # Get file bytes and compute a hash
         bytes_data = uploaded_file.getvalue()
-        with open(uploaded_file.name, "wb") as file:
-            file.write(bytes_data)
-        st.image(uploaded_file, caption="Uploaded Image", use_container_width=True)
+        file_hash = hashlib.sha256(bytes_data).hexdigest()
         
-        # Initialize session state if not present.
-        if "scenario" not in st.session_state:
+        # Reset session state only if the file content has changed
+        if ("last_uploaded_hash" not in st.session_state) or (st.session_state.last_uploaded_hash != file_hash):
             st.session_state.scenario = None
-        if "story" not in st.session_state:
             st.session_state.story = None
-        if "audio_data" not in st.session_state:
             st.session_state.audio_data = None
-
+            st.session_state.last_uploaded_hash = file_hash
+        
+        # Save the uploaded file locally.
+        with open(uploaded_file.name, "wb") as file:
+            file.write(bytes_data)
+        st.image(uploaded_file, caption="Uploaded Image", use_container_width=True)
+        
         # Stage 1: Image to Text
         if st.session_state.scenario is None:
             st.text("Processing img2text...")
             st.session_state.scenario = img2text(uploaded_file.name)
         st.write(st.session_state.scenario)
         
-        # Stage 2: Text to Story (only if not generated or after regeneration)
+        # Stage 2: Text to Story
         if st.session_state.story is None:
             st.text("Generating a story...")
             st.session_state.story = text2story(st.session_state.scenario)
-            st.session_state.audio_data = text2audio(st.session_state.story)
         st.write(st.session_state.story)
         
-        # Two buttons: "Play Audio" and "Story regeneration"
-        col1, col2 = st.columns(2)
-        
-        with col1:
-            if st.button("Play Audio"):
-                st.audio(
-                    st.session_state.audio_data["audio"],
-                    format="audio/wav",
-                    start_time=0,
-                    sample_rate=st.session_state.audio_data["sampling_rate"]
-                )
+        # Stage 3: Story to Audio data
+        if st.session_state.audio_data is None:
+            st.text("Generating audio data...")
+            st.session_state.audio_data = text2audio(st.session_state.story)
         
-        with col2:
-            if st.button("Story regeneration"):
-                st.session_state.story = text2story(st.session_state.scenario)
-                st.session_state.audio_data = text2audio(st.session_state.story)
-                st.write("New Story:")
-                st.write(st.session_state.story)
+        # Play Audio button – uses stored audio_data.
+        if st.button("Play Audio"):
+            st.audio(
+                st.session_state.audio_data["audio"],
+                format="audio/wav",
+                start_time=0,
+                sample_rate=st.session_state.audio_data["sampling_rate"]
+            )
 
 if __name__ == "__main__":
     main()