YouTube summariser

Title:

Youtube Summarizer

Project Guide:Prof. Shruti Agarwal

Start

Index

PROBLEM STATEMENT

INTRODUCTION

LITERATURE SURVEY

TEAM

PROPOSED METHOD

MAP

IMAGE

METHODOLOGY

THANKS

CONCLUSION

Group 3

TEAM

Vaishnavi Pinagale

Atharvaa Rane

Mrunal Golatkar

Surabhi Pathare

INFT-A19103A0033

INFT-A191o1A0012

INFT-A19101A0010

INFT-A19101A0076

+ info

+ info

+ info

+ info

Abstract

• Enormous number of video recordings are being created and shared on the YouTube throughout the day.
• It becomes really difficult to spend time in watching such videos which may have a longer duration than expected and sometimes our efforts may become futile if we aren’t able to find relevant information out of it.
• Summarizing transcripts of such videos will allows us to quickly look out for the important patterns in the video and helps us to save time and efforts to go through the whole content of the video.

Problem Statement

The proposed system provides information about the smart city GUI application that can be browsed from anywhere using internet service with all details of a particular city that has to be visited

Literature Survey

• https://www.crio.do/projects/python-youtube-transcript/
• https://ieeexplore.ieee.org/document/9327918
• In the first link mentioned below, the summarizer can only summarize videos which are in English language and have transcripts by default. But in our project, the user will be able to provide link of a non-English video which will be summarized in English . Also, if the transcripts are not available by default, the same can be obtained the transcripts by the audio file of the video by speech recognition .

Reference

• https://www.crio.do/projects/python-youtube-transcript/
• https://pypi.org/project/youtube-transcript-api/
• https://www.thepythoncode.com/article/text-summarization-using-huggingface-transformers-python
• https://pythonbasics.org/transcribe-audio/
• https://www.thepythoncode.com/article/translate-text-in-python

Screenshot

Software Requirements

• Python
• GUI using tkinter

Libraries:

1.tkinter (For GUI)2.youtube_transcript_api (to obtain transcripts of the videos for which transcripts are available) 3.pafy-(to retrieve video id of the link provided) 4.json 5.torch 6.transformers (pretrained model for text summarizarion) 7.pytube ( to get details of the youtube video) 8.moviepy(for downloading the yt video) 9.ibm_watson(for converting youtube video to text format) 10.smptlib-(to send mail)

Working

1. mp3 to wav conversion
2. loading the audio file
3. feeding the audio file to a speech recognition system.

• HuggingFace transformers(torch) -for text summarization of the transcript obtained.
• googletrans- language translator using google API for non-english videos.

Modules

1. Building a GUI
2. Get transcript for a given video
3. Performing text summarization
4. Displaying summarized text
5. Emailing the summarizer text

System Architecture

The user will have to paste the link of the YouTube video to be summarized and click on summarize.

Through the YouTube video id or audio of the video, obtaining the transcripts.

Summarizing the transcripts using NLP techniques.

Displaying the summarized text on the GUI.

Emailing the summariser text to the user

Timeline

Automation

Transcribing audio using Speech Recognition

Result & Discusssion

From the YouTube link if transcripts are available then we retrieve the transcripts and then summarize the video.

If the transcripts are not available then the video is converted to text format and then it is summarized.

The Summary will be displayed to the screen and it will be mailed to the specified email id.

Conclusion

Hence the smart city project provides information regarding the various aspects of the city such as places to visit,maps etc. The implementation of this project solves the problem of finding different places and hotels for the new visitors.

THANKS!

Design