Converting youtube video into text for article

If you own a blog/website, you may wonder what is the easiest way to fill your blog or website with articles. You can get a help from article writer from website like fiverr, odesk or freelancer.com, and although most of the writer offer competitive rate with good quality, it still cost you money. You may try to write the article by yourself, but it’s not gonna be easy if you are not a native English speaker. On this article, I will show you the way where you can create articles for free and with minimal effort : Converting youtube video into text for article.





Disclaimer!! It may be illegal to transcribe youtube into text in your country. Do your research in regard of this law and do it at your own risk!

 

Step 1 : Find the target youtube video to be converted into text

This is, of course, the first step you need to do in order for you for Converting youtube video into text for article. To make sure that your article will have a decent quality, avoid following video :

  • Vlog video
  • Video with low quality audio oputput
  • Video in which involve non-native speaker with funny accent
  • Video with outside noise

The reason is, most of above type video will provide low quality transcriptions, and you will end up having to fix many errors from the output text. I recommend using review-type video which the narrator have a clear voice in English, no accent and about 5- 10 minutes.

We don’t want to use long video of more than 10 minutes because the amount of text you have to proofread will be substantially bigger. Remember, we wan’t to make it as easy as possible.




 

Step 2 : Use google chrome and Turn on the CC/Subtitle button

YouTube will automatically creates subtitles or closed captions when someone upload video. Youtube, which is now own by Google, uses speech recognition technology to transcribe the speech portion of the video into closed captions that will be displayed when user click on the CC button (see screenshot).

 

Converting youtube video into text for article
Youtube closed caption button

Youtube will only creates text transcript/subtitle if the video has decent audio quality and not too many outside noise. This transcript, is hidden inside inside obfuscated JavaScript, this hidden file is what we are going to unlock for Converting youtube video into text for article.

Step 3 : Find and copy the audio transcript from youtube

Here’s step by step guide on how to find text audio transcript from youtube :

  1. Open the target video on http://www.youtube.com/ that you want to turn into article
  2. Make sure the video has the “CC” or subtitle button, click on it if it has.
  3. Assuming you are using windows and google chrome, press F12 from your keyboard (Fn + F12 for some laptop).  If you are using Mac, press Option+Cmd+J.
  4. Pressing that button will open up javascript console :
  5. Converting youtube video into text for article
  6. Click on console tab
  7. Copy and paste below code, and press enter
  8.  

    if(yt.config_.TTS_URL.length) window.location.href=yt.config_.TTS_URL+"&kind=asr&fmt=srv1&lang=en"
    
    
  9. This will open up below files. This is the full audio transcribe that we want to use for article.Converting youtube video into text for article
  10. Copy all of the XML file and paste it into words
  11. To remove XML formatting in word, click CTRL + H. It will bring out Replace dialog.
  12. On the find what field, type or copy paste this :  \<*\>
  13. Click “Use wildcard”
  14. Converting youtube video into text for article
  15. Click “Replace All”
  16. Now you should have clean text, but there is usually &#39; which represent single quote, we need to clean that also
  17. So put this :  &#39;  in find what, and in Replace type or copy paste this:
  18. The last step is manual step, you need to remove new line and use comma, paragraph etc.
  19. Converting youtube video into text for article

 

There you are, you now have one article for your web/blog content taken from youtube audio. Keep in mind that youtube automatic transcription is not 100% perfect, you still have to proofread your article in case there are grammatical errors or wrong interpretations of word. Do this and your blog will have lots of content in no time.

 




 

Leave a Reply

Your email address will not be published.

*