Create app.py

app.py

@@ -0,0 +1,47 @@
+import streamlit as st
+from transformers import pipeline
+from pytesseract import image_to_string
+import pyttsx3
+
+# Initialize text summarization pipeline and text-to-speech engine
+summarizer = pipeline("summarization")
+engine = pyttsx3.init()
+
+def summarize_pdf(pdf_file):
+    """Extracts text from PDF, summarizes it, and returns discussion points."""
+
+    # Extract text from PDF using pytesseract (consider alternatives if performance is critical)
+    text = image_to_string(pdf_file.read(), config='--psm 6')  # Improve accuracy for complex layouts
+
+    # Summarize the extracted text
+    summary = summarizer(text, max_length=150, min_length=50, do_sample=False)  # Control summary length
+    discussion_points = [point["summary_text"] for point in summary["sentences"]]  # Extract discussion points
+
+    return discussion_points
+
+def play_audio(text):
+    """Converts text to speech and plays it."""
+
+    engine.say(text)
+    engine.runAndWait()
+
+def main():
+    """Streamlit app layout and functionality."""
+
+    st.title("Discussion Point Summarizer")
+    st.subheader("Upload a PDF to generate key discussion points.")
+
+    uploaded_file = st.file_uploader("Choose a PDF file", type="pdf")
+
+    if uploaded_file is not None:
+        discussion_points = summarize_pdf(uploaded_file)
+
+        st.header("Discussion Points:")
+        for point in discussion_points:
+            st.write(f"- {point}")
+
+        if st.button("Listen to Discussion Points"):
+            play_audio(" ".join(discussion_points))  # Combine points for smoother speech
+
+if __name__ == "__main__":
+    main()