app.py

# https://huggingface.co/spaces/Mishmosh/MichelleAssessment3
#!pip install PyPDF2
#!pip install sentencepiece
#!pip install pdfminer.six
#!pip install pdfplumber
#!pip install pdf2image
#!pip install Pillow
#!pip install pytesseract
# @title
!apt-get install poppler-utils
!apt install tesseract-ocr
!apt install libtesseract-dev
import PyPDF2
from pdfminer.high_level import extract_pages, extract_text
from pdfminer.layout import LTTextContainer, LTChar, LTRect, LTFigure
import pdfplumber
from PIL import Image
from pdf2image import convert_from_path
import pytesseract
import os
def text_extraction(element):
    # Extracting the text from the in-line text element
    line_text = element.get_text()

    # Find the formats of the text
    # Initialize the list with all the formats that appeared in the line of text
    line_formats = []
    for text_line in element:
        if isinstance(text_line, LTTextContainer):
            # Iterating through each character in the line of text
            for character in text_line:
                if isinstance(character, LTChar):
                    # Append the font name of the character
                    line_formats.append(character.fontname)
                    # Append the font size of the character
                    line_formats.append(character.size)
    # Find the unique font sizes and names in the line
    format_per_line = list(set(line_formats))

    # Return a tuple with the text in each line along with its format
    return (line_text, format_per_line)
# @title
# Create a function to crop the image elements from PDFs
def crop_image(element, pageObj):
    # Get the coordinates to crop the image from the PDF
    [image_left, image_top, image_right, image_bottom] = [element.x0,element.y0,element.x1,element.y1]
    # Crop the page using coordinates (left, bottom, right, top)
    pageObj.mediabox.lower_left = (image_left, image_bottom)
    pageObj.mediabox.upper_right = (image_right, image_top)
    # Save the cropped page to a new PDF
    cropped_pdf_writer = PyPDF2.PdfWriter()
    cropped_pdf_writer.add_page(pageObj)
    # Save the cropped PDF to a new file
    with open('cropped_image.pdf', 'wb') as cropped_pdf_file:
        cropped_pdf_writer.write(cropped_pdf_file)

# Create a function to convert the PDF to images
def convert_to_images(input_file,):
    images = convert_from_path(input_file)
    image = images[0]
    output_file = "PDF_image.png"
    image.save(output_file, "PNG")

# Create a function to read text from images
def image_to_text(image_path):
    # Read the image
    img = Image.open(image_path)
    # Extract the text from the image
    text = pytesseract.image_to_string(img)
    return text
# @title
# Extracting tables from the page

def extract_table(pdf_path, page_num, table_num):
    # Open the pdf file
    pdf = pdfplumber.open(pdf_path)
    # Find the examined page
    table_page = pdf.pages[page_num]
    # Extract the appropriate table
    table = table_page.extract_tables()[table_num]
    return table

# Convert table into the appropriate format
def table_converter(table):
    table_string = ''
    # Iterate through each row of the table
    for row_num in range(len(table)):
        row = table[row_num]
        # Remove the line breaker from the wrapped texts
        cleaned_row = [item.replace('\n', ' ') if item is not None and '\n' in item else 'None' if item is None else item for item in row]
        # Convert the table into a string
        table_string+=('|'+'|'.join(cleaned_row)+'|'+'\n')
    # Removing the last line break
    table_string = table_string[:-1]
    return table_string
# @title
def read_pdf(pdf_path):
  # create a PDF file object
  pdfFileObj = open(pdf_path, 'rb')
  # create a PDF reader object
  pdfReaded = PyPDF2.PdfReader(pdfFileObj)

  # Create the dictionary to extract text from each image
  text_per_page = {}
  # We extract the pages from the PDF
  for pagenum, page in enumerate(extract_pages(pdf_path)):
      print("Elaborating Page_" +str(pagenum))
      # Initialize the variables needed for the text extraction from the page
      pageObj = pdfReaded.pages[pagenum]
      page_text = []
      line_format = []
      text_from_images = []
      text_from_tables = []
      page_content = []
      # Initialize the number of the examined tables
      table_num = 0
      first_element= True
      table_extraction_flag= False
      # Open the pdf file
      pdf = pdfplumber.open(pdf_path)
      # Find the examined page
      page_tables = pdf.pages[pagenum]
      # Find the number of tables on the page
      tables = page_tables.find_tables()


      # Find all the elements
      page_elements = [(element.y1, element) for element in page._objs]
      # Sort all the elements as they appear in the page
      page_elements.sort(key=lambda a: a[0], reverse=True)

      # Find the elements that composed a page
      for i,component in enumerate(page_elements):
          # Extract the position of the top side of the element in the PDF
          pos= component[0]
          # Extract the element of the page layout
          element = component[1]

          # Check if the element is a text element
          if isinstance(element, LTTextContainer):
              # Check if the text appeared in a table
              if table_extraction_flag == False:
                  # Use the function to extract the text and format for each text element
                  (line_text, format_per_line) = text_extraction(element)
                  # Append the text of each line to the page text
                  page_text.append(line_text)
                  # Append the format for each line containing text
                  line_format.append(format_per_line)
                  page_content.append(line_text)
              else:
                  # Omit the text that appeared in a table
                  pass

          # Check the elements for images
          if isinstance(element, LTFigure):
              # Crop the image from the PDF
              crop_image(element, pageObj)
              # Convert the cropped pdf to an image
              convert_to_images('cropped_image.pdf')
              # Extract the text from the image
              image_text = image_to_text('PDF_image.png')
              text_from_images.append(image_text)
              page_content.append(image_text)
              # Add a placeholder in the text and format lists
              page_text.append('image')
              line_format.append('image')

          # Check the elements for tables
          if isinstance(element, LTRect):
              # If the first rectangular element
              if first_element == True and (table_num+1) <= len(tables):
                  # Find the bounding box of the table
                  lower_side = page.bbox[3] - tables[table_num].bbox[3]
                  upper_side = element.y1
                  # Extract the information from the table
                  table = extract_table(pdf_path, pagenum, table_num)
                  # Convert the table information in structured string format
                  table_string = table_converter(table)
                  # Append the table string into a list
                  text_from_tables.append(table_string)
                  page_content.append(table_string)
                  # Set the flag as True to avoid the content again
                  table_extraction_flag = True
                  # Make it another element
                  first_element = False
                  # Add a placeholder in the text and format lists
                  page_text.append('table')
                  line_format.append('table')

                  # Check if we already extracted the tables from the page
                  if element.y0 >= lower_side and element.y1 <= upper_side:
                      pass
                  elif not isinstance(page_elements[i+1][1], LTRect):
                      table_extraction_flag = False
                      first_element = True
                      table_num+=1


      # Create the key of the dictionary
      dctkey = 'Page_'+str(pagenum)
      # Add the list of list as the value of the page key
      text_per_page[dctkey]= [page_text, line_format, text_from_images,text_from_tables, page_content]

  # Closing the pdf file object
  pdfFileObj.close()

  # Deleting the additional files created
  #os.remove('cropped_image.pdf')
  #os.remove('PDF_image.png')
  return text_per_page

#google drive
#from google.colab import drive
#drive.mount('/content/drive')
#read PDF

pdf_path = 'https://huggingface.co/spaces/Mishmosh/MichelleAssessment3/blob/main/Article%2011%20Hidden%20Technical%20Debt%20in%20Machine%20Learning%20Systems.pdf' #article 11

text_per_page = read_pdf(pdf_path)

# This section finds the abstract. My plan was to find the end of the abstract by identifying the same font size as the text 'abstract', but it was too late
#to try this here since the formatting of the text has already been removed.
#  Instead I extracted just one paragraph. If an abstract is more than 1 paragraph this will not extract the entire abstract
abstract_from_pdf='' # define empty variable that will hold the text from the abstract
found_abstract=False # has the abstract been found
for key in text_per_page.keys(): # go through keys in dictionary
  current_item=text_per_page[key] #current key
  for paragraphs in current_item: #go through each item
    for index,paragraph in enumerate(paragraphs): #go through each line
      if 'Abstract\n' == paragraph: #does line match paragraph
        found_abstract=True #word abstract has been found
        abstract_from_pdf=paragraphs[index+1] #get next paragraph
    if found_abstract: #if abstract found
      break
print(abstract_from_pdf)

from transformers import pipeline
summarizer = pipeline("summarization", model="ainize/bart-base-cnn")
#summarizer = pipeline("summarization", model="linydub/bart-large-samsum") # various models were tried and the best one was selected
#summarizer = pipeline("summarization", model="slauw87/bart_summarisation")
#summarizer = pipeline("summarization", model="facebook/bart-large-cnn")
#summarizer = pipeline("summarization", model="google/pegasus-cnn_dailymail")
#print(summarizer(abstract_from_pdf, max_length=50, min_length=5, do_sample=False))
summarized_text=(summarizer(abstract_from_pdf))
print(summarized_text)
#summary_of_abstract=str(summarizer)
#type(summary_of_abstract)
#print(summary_of_abstract)

# the aim of this section of code is to get a summary of just one sentence by summarizing the summary all while the summary is longer than one sentence.
# unfortunately, I tried many many models and none of them actually summarize the text to as short as one sentence.
#I had searched for ways to fine tune the summarization model to specify that the summarization should be done in just one sentence but did not find a way to implement it
from transformers import pipeline
summarized_text_list_list=summarized_text_list['summary_text']
summarizer = pipeline("summarization", model="facebook/bart-large-cnn")
#print(summarizer)
number_of_sentences=summarized_text_list_list.count('.')
print(number_of_sentences)
while(number_of_sentences)>1:
    print(number_of_sentences)
    summarized_text_list_list=summarizer(summarized_text_list_list)[0]['summary_text']
    number_of_sentences-=1
    print(summarized_text_list_list)
    print(number_of_sentences)


#text to speech
!pip install git+https://github.com/huggingface/transformers.git
!pip install datasets sentencepiece
import torch
import soundfile as sf
from IPython.display import Audio
from datasets import load_dataset
from transformers import pipeline
from transformers import SpeechT5Processor, SpeechT5ForTextToSpeech
processor = SpeechT5Processor.from_pretrained("microsoft/speecht5_tts")
model = SpeechT5ForTextToSpeech.from_pretrained("microsoft/speecht5_tts")
#text = "The future belongs to those who believe in the beauty of their dreams."
#text = (summarized_text_list_list)

inputs = processor(text=summarized_text_list_list, return_tensors="pt")
from datasets import load_dataset
embeddings_dataset = load_dataset("Matthijs/cmu-arctic-xvectors", split="validation")

import torch
speaker_embeddings = torch.tensor(embeddings_dataset[7306]["xvector"]).unsqueeze(0)
spectrogram = model.generate_speech(inputs["input_ids"], speaker_embeddings)
from transformers import SpeechT5HifiGan
vocoder = SpeechT5HifiGan.from_pretrained("microsoft/speecht5_hifigan")
with torch.no_grad():
    speech = vocoder(spectrogram)
speech = model.generate_speech(inputs["input_ids"], speaker_embeddings, vocoder=vocoder)
Audio(speech, rate=16000)