app.py

from flask import Flask, Response, request, jsonify, render_template
from final_summary import *
from Meeting_duration import *
from keywords import *
from PyPDF2 import PdfReader
import PyPDF2
from transformers import pipeline
import docx
import webvtt
import csv
import pandas as pd
import io
import random
import string
import os
import re
import speech_recognition as sr
import os
from pydub import AudioSegment
from pydub.silence import split_on_silence
from moviepy.editor import VideoFileClip
import shutil
from audio_to_text import *


app = Flask(__name__)

@app.route('/')
def home():
    return render_template('index.html')

@app.route('/summary_page',methods=['GET','POST'])
def summary_page():
    return render_template('summary_page.html')


@app.route('/attendees_page',methods=['GET','POST'])
def attendees_page():
    return render_template('attendees_page.html')

meet_attendee_list = []
@app.route('/attendees_result',methods=['GET','POST'])
def attendees_result():
    if request.method == 'POST':
        transcript = request.files['myfile']
        transcript.save(transcript.filename)

        with open(transcript.filename, 'r') as f:
            vtt_lines = f.readlines()
        timestamps = []
        for line in vtt_lines:
            if regex.match(r'^\d{2}:\d{2}:\d{2}.\d{3} --> \d{2}:\d{2}:\d{2}.\d{3}$', line.strip()):
                timestamps.append(line.strip())

        formatted_timestamps = []
        for ts in timestamps:
            start, end = ts.split(' --> ')
            start = datetime.strptime(start, '%H:%M:%S.%f')
            end = datetime.strptime(end, '%H:%M:%S.%f')
            formatted_timestamps.append((start, end))

        # print(type(end))
        # print(start,"----",end)
        # print(formatted_timestamps)

        lenn = len(formatted_timestamps) 
        # print(formatted_timestamps[lenn - 1])

        dt = formatted_timestamps[lenn - 1]

        last_len = len(dt)

        aaa = dt[last_len - 1]
        aaa
        minutes = aaa.minute

        hours = aaa.hour

        secs = aaa.second

        f = open(transcript.filename, "r")
        string = f.read()
        # Stores the indices and attendees
        dels = []
        namee = []

        # ans = ""
        for i in range(len(string)):
            
            # If opening delimiter
            # is encountered
            if (string[i] == '<') :
                dels.append(i + 2)

            # If closing delimiter
            # is encountered
            elif (string[i] == '>' and len(dels) != 0):

                # Extract the position
                # of opening delimiter
                pos = dels[-1]
                dels.pop()

                # Length of substring
                length = i - 1 - pos

                # Extract the substring
                ans = string[pos + 1 : pos + 1 + length]
                
                #replacing blank space between names with _
                names = ans.replace(" ", "_")
                
                namee.append(names)
                
        meet_attendees = set(namee)
        meet_attendees.remove("")  

        global meet_attendee_list
        meet_attendee_list = list(meet_attendees)

        S = ', '.join(meet_attendees)
        # file = open("Active_Attendees.txt","w")

        # file.write("Total Active attendees: ")
        # file.write(str(len(meet_attendees)))

        # file.write("\nTotal Active attendees Name: \n")

        # file.write(str(S))
        # file.close()

        # ***************************Download file********************
        # column_name = "Active_Attendees"
        # with open('example.csv', mode='a', newline='') as file:

        #     # Create a writer object
        #     writer = csv.writer(file)

        #     # Write the data to the CSV file
        #     writer.writerow([column_name])

        
        # df = pd.read_csv('example.csv')
        # # Create a new list of values to insert into the column
        # new_column_data = list(meet_attendees)

        # # Append the new data to the desired column
        # df[column_name] = df[column_name].append(pd.Series(new_column_data))

        # # Write the updated DataFrame to a new CSV file
        # df.to_csv('updated_example.csv', index=False)
        # **********************

    return render_template('attendees_result.html',
                            hour = hours,
                            minutes = minutes,
                            sec = secs,
                            meet_attendees_name =str(S),                           
                            total_att = len(meet_attendees),
                            file_name = transcript.filename
                           )

# Route for TextField Summarization
@app.route('/summarize_text',methods=['POST'])
def summarize_text():
    if request.method == 'POST':
        text_summary = request.form['text_summary']

        Summary = generate_summary(text_summary)

        #convert list into string
        global summary_string
        summary_string = ' '.join(Summary)   

        # **********Count total words and sentences***************
        # text = "This is a sample sentence. Here is another one!"
            # Count the number of words
        words = re.findall(r'\w+', summary_string)
        num_words = len(words)

            # Count the number of sentences
        sentences = re.findall(r'[^\s][^.!?]*[.!?]', summary_string)
        num_sentences = len(sentences)

        Keywords = keywords(text_summary)

    return render_template('summary_result.html', 
                           text_summary=summary_string,
                           total_words = num_words,
                           total_sentences = num_sentences,
                           top_keywords = Keywords)

# Route for File upload Summarization
summary_string = ""
@app.route('/summarize',methods=['POST'])
def summarize():
    
    if request.method == 'POST':
        
        #fetching file from html form
        text = request.files['fileUpload']
        text.save(text.filename)

        #for getting file extension
        find_file = text.filename
        x = find_file.split(".")
        extension = x[1]


        if extension == "pdf":
            reader = PdfReader(text.filename)

            page = reader.pages[0] #for single page
           
            file_text = page.extractText()

        elif extension == "docx":

            doc = docx.Document(text.filename)

            # Extract the text from the document
            file_text = '\n'.join([paragraph.text for paragraph in doc.paragraphs])

        elif extension == "vtt":
            # Load VTT file
            captions = webvtt.read(text.filename)

            # Iterate through captions and extract text
            file_text = ''
            for caption in captions:
                file_text += caption.text + '\n'
            # print("Original text after removing stopwords =========>>>>  ",file_text)
            
            Keywords = keywords(file_text)

        else:
            #for txt file
            f = open(text.filename, "r", encoding="utf8")
            file_text = f.read()
            Keywords = keywords(file_text)

        Summary = generate_summary(file_text)

        #convert list into string
        global summary_string
        summary_string = ' '.join(Summary)   

        # **********Count total words and sentences***************
        # text = "This is a sample sentence. Here is another one!"
            # Count the number of words
        words = re.findall(r'\w+', summary_string)
        num_words = len(words)

            # Count the number of sentences
        sentences = re.findall(r'[^\s][^.!?]*[.!?]', summary_string)
        num_sentences = len(sentences)
        # ***************************************************** 
    return render_template('summary_result.html',
                               text_summary=summary_string,
                               file_name = text.filename,
                               total_words = num_words,
                               total_sentences = num_sentences,
                               top_keywords = Keywords)

        # ******************END************************



@app.route('/download_csv')
def download_csv():
    
    
    insert_data = ["Active_attendees"]
    insert_data.extend(meet_attendee_list)

    insert_names= [[x] for x in insert_data]

    # Create a buffer to write the CSV data to
    output = io.StringIO()

    # Use the csv module to write the data to the buffer
    writer = csv.writer(output)
    for row in insert_names:
        writer.writerow(row)

    # Set the appropriate response headers
    response = Response(output.getvalue(), mimetype='text/csv')

    # *******Generate random file name***
    # Generate a sequence of 8 random characters (4 letters and 4 numbers)
    random_chars = ''.join(random.choices(string.ascii_lowercase + string.digits, k=16))

    # Insert an underscore after every 4 characters
    result = '_'.join([random_chars[i:i+4] for i in range(0, len(random_chars), 4)])
    file_name = "Attendance_" + result +".csv"
 
    # ************************
    response.headers.set("Content-Disposition", "attachment", filename=file_name)

    return response


# ***********************Download Summary***************************
@app.route('/download_summary',methods=['POST'])
def download_summary():

    if request.method == 'POST':
        text_summary = request.form['text_summary']
    # Add a summary in the file
    
    # Delete a file
        os.remove('summary.txt')

        with open('summary.txt', 'a') as f:
            f.write(text_summary + '\n')
        # return 'Data added successfully!'

        with open('summary.txt', 'r') as f:
            data = f.read()
        return Response(
            data,
            mimetype="text/plain",
            headers={"Content-disposition":
                    "attachment; filename=summary.txt"})
# ****************************************************************


# ***********************Voice to Text Page***************************
@app.route('/voice_to_text')
def voice_to_text():
    return render_template('voice_to_text.html')
# ****************************************************************

# ***********************Voice to Text Result Page****************************
@app.route('/voice_to_text_result', methods=['GET', 'POST'])
def voice_to_text_result():
    r = sr.Recognizer()
    if request.method == 'POST':
        video_file = request.files['fileUpload']
        video_file_name = video_file.filename
        
        video_folder_name = "video_chunks"
        if not os.path.isdir(video_folder_name):
            os.mkdir(video_folder_name)

        target_folder = "video_chunks"
        video_file.save(os.path.join(target_folder, video_file.filename))
       
        # *************Video to audio*************8
        # clip = VideoFileClip("./Input_Files/sample_video.mp4")
        path = "./video_chunks/" + video_file_name

        clip = VideoFileClip(path)

        # Extract the audio from the video
        audio = clip.audio

        # *******Generate random file name***
        # Generate a sequence of 8 random characters (4 letters and 4 numbers)
        random_chars1 = ''.join(random.choices(string.ascii_lowercase + string.digits, k=16))

        # Insert an underscore after every 4 characters
        result = '_'.join([random_chars1[i:i+4] for i in range(0, len(random_chars1), 4)])
        video_file_name_id = "video_to_audio" + result +".wav"

        # ***************************************

        # Save the audio to a WAV file
        audio.write_audiofile(video_file_name_id, codec='pcm_s16le') 

        # """Split audio into chunks and apply speech recognition"""
    # Open audio file with pydub
        whole_text = convert_audio_to_text(video_file_name_id)

        os.remove(video_file_name_id)
        shutil.rmtree("./audio-chunks")
        # shutil.rmtree("./video_chunks")

    # Return text for all chunks

    return render_template('voice_result.html', whole_text= whole_text)
    
# ****************************************************************

@app.route('/audio_to_text', methods=['GET', 'POST'])
def audio_to_text():
    r = sr.Recognizer()
    if request.method == 'POST':
        audio_file = request.files['fileUpload']
        audio_file_name = audio_file.filename
        
        whole_text = convert_audio_to_text(audio_file_name)
        shutil.rmtree("./audio-chunks")
        # shutil.rmtree("./video_chunks")

    # Return text for all chunks

    return render_template('voice_result.html', whole_text= whole_text)

if __name__ == '__main__':
    app.run(debug=True, port=5000, threaded=True)