# To read the PDF 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 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 hf_name = 'pszemraj/led-large-book-summary' summarizer = pipeline( "summarization", hf_name, device=0 if torch.cuda.is_available() else -1, ) 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) 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 # 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() return text_per_page def upload_file(files): print("here") file_paths = [file.name for file in files] return file_paths with gr.Blocks() as demo: file_output = gr.File() upload_button = gr.UploadButton("Click to Upload a File", file_types=[".pdf"]) upload_button.upload(upload_file, upload_button, file_output) pdf_path = file_output demo.launch(debug=True) text_per_page = read_pdf(pdf_path) Page_0 = text_per_page['Page_0'] def nested_list_to_string(nested_list): result = '' for element in nested_list: if isinstance(element, list): # Check if the element is a list result += nested_list_to_string(element) # Recursively process the list elif isinstance(element, str): # Check if the element is a string result += element # Append the string to the result return result Page_0 = text_per_page['Page_0'] string_result = nested_list_to_string(Page_0) def extract_abstract(page_0): def nested_list_to_string(nested_list): result = '' for element in nested_list: if isinstance(element, list): # Check if the element is a list result += nested_list_to_string(element) # Recursively process the list elif isinstance(element, str): # Check if the element is a string result += element # Append the string to the result return result # Convert the nested list into a single string full_text = nested_list_to_string(page_0) # Find the start of the 'Abstract' section and the end of it (start of 'Introduction') start_index = full_text.find('Abstract') end_index = full_text.find('Introduction') # If both 'Abstract' and 'Introduction' are found, extract the text in between if start_index != -1 and end_index != -1: # Extract the text and remove the word 'Abstract' abstract_text = full_text[start_index + len('Abstract'):end_index] return abstract_text.strip() else: return "Abstract or Introduction section not found." # Example usage Page_0 = text_per_page['Page_0'] abstract_text = extract_abstract(Page_0) wall_of_text = abstract_text result = summarizer( wall_of_text, min_length=1, max_length=30, no_repeat_ngram_size=3, encoder_no_repeat_ngram_size=3, repetition_penalty=3.5, num_beams=4, early_stopping=True, ) # Access the first element of the list (which is the dictionary) and then the value of 'summary_text' summary_string = result[0]['summary_text'] print(summary_string) from transformers import SpeechT5Processor, SpeechT5ForTextToSpeech processor = SpeechT5Processor.from_pretrained("microsoft/speecht5_tts") model = SpeechT5ForTextToSpeech.from_pretrained("microsoft/speecht5_tts")