app.py

import gradio as gr
from pptx import Presentation
from pptx.util import Inches
import pdfplumber
from transformers import pipeline
import tempfile

# Initialize the summarization model
summarizer = pipeline("summarization", model="facebook/bart-large-cnn")

# Function to process PDF and create PowerPoint
def pdf_to_pptx(pdf_file):
    # Step 1: Extract text from the PDF
    pdf_text = ""
    with pdfplumber.open(pdf_file) as pdf:
        for page in pdf.pages:
            pdf_text += page.extract_text()

    # Step 2: Summarize text
    def chunk_text(text, max_len=1024):
        return [text[i:i+max_len] for i in range(0, len(text), max_len)]
    
    chunked_text = chunk_text(pdf_text)
    summarized_text = " ".join([summarizer(chunk)[0]['summary_text'] for chunk in chunked_text])

    # Step 3: Create PowerPoint presentation
    presentation = Presentation()
    slide_content = summarized_text.split(". ")

    for i, content in enumerate(slide_content):
        slide = presentation.slides.add_slide(presentation.slide_layouts[1])  # Title and Content layout
        title = slide.shapes.title
        body = slide.placeholders[1]
        title.text = f"Slide {i+1}"  # Slide title
        body.text = content  # Slide content

    # Step 4: Save presentation to a temporary file and return path
    temp_pptx = tempfile.NamedTemporaryFile(delete=False, suffix=".pptx")
    presentation.save(temp_pptx.name)
    return temp_pptx.name

# Gradio app interface
def generate_pptx(pdf_file):
    pptx_file_path = pdf_to_pptx(pdf_file.name)
    return pptx_file_path

# Define Gradio interface
interface = gr.Interface(
    fn=generate_pptx,
    inputs=gr.File(label="Upload PDF File"),
    outputs=gr.File(label="Download PowerPoint Presentation"),
    title="PDF to PowerPoint Converter",
    description="Upload a PDF file to convert it into a professional PowerPoint presentation."
)

# Launch the app
interface.launch()