!pip install PyPDF2
import PyPDF2
from PyPDF2 import PdfReader

# Open the PDF file
pdf_file_path = "C:/Users/styli/OneDrive/Desktop/OPIT/AAI/Assessment 3/Article 11 Hidden Technical Debt in Machine Learning Systems.pdf"

# Use the new PdfReader class to read the PDF
pdf_reader = PdfReader(pdf_file_path)

# Initialize an empty string to hold the content
content = ""

# Extract text from each page and concatenate it to the content variable
for page in pdf_reader.pages:
    text = page.extract_text()
    content += text + "\n"  
import re

# Pattern to match in-text citations (e.g., [1], [2, 3], or (Author, Year))
in_text_citations_pattern = r'\[\d+(,\s*\d+)*\]|\([A-Za-z ]+,\s*\d{4}\)'

# Pattern to match the References section at the end of the document
references_section_pattern = r'\nReferences[\s\S]*'

# Remove in-text citations
text_without_citations = re.sub(in_text_citations_pattern, '', content)

# Remove the References section
text_without_references = re.sub(references_section_pattern, '', text_without_citations)

# The new variable without the references
cleaned_text = text_without_references
!pip install transformers
from transformers import T5ForConditionalGeneration, T5Tokenizer

model_name = 't5-large'
tokenizer = T5Tokenizer.from_pretrained(model_name)
model = T5ForConditionalGeneration.from_pretrained(model_name)
# Encode the text to input_ids (token ids)
to_tokenize = "summarize: " + content
inputs = tokenizer.encode(to_tokenize, return_tensors="pt", max_length=512, truncation=True)
summary_ids = model.generate(inputs, max_length=48, length_penalty=2, num_beams=5, early_stopping=True)
summary = tokenizer.decode(summary_ids[0], skip_special_tokens=True)
Summary=summary.capitalize()
print(Summary)
pip install pyttsx3
import pyttsx3
from IPython.display import Audio

# Setup the text-to-speech engine
engine = pyttsx3.init()


# Generate speech
engine.save_to_file(Summary, 'summary.mp3')
engine.runAndWait()  # Blocks while processing all currently queued commands

# Play the audio in Jupyter Notebook
Audio('summary.mp3', autoplay=True)