We are going to learn how to convert an audio to text using YouTube as an audio to text converter. We are going to use the free transcription service of YouTube to convert an audio file to text.
I write for people. So as freelancer i always had to take reference from different sources like YouTube to get information on the matters that i didn’t know about. These writing specific websites always check for duplicate content using copyscape and other such kind of tools. Now we at hackstories are very lazy people and we are always looking for ways through which we can cheat without getting caught.
We had an idea of using YouTube videos as a source of writing articles. For example if the article i had to write was “How to transcribe audio to text?” I would go to YouTube and search for that exact term. As the maker of that video spoke about the content of that video i used to write it all down on my notepad. Pretty crude huh? Then i realized that there has to be a better way to handle this. I researched and found that YouTube provides a service called transcriptions in English, which are auto-generated.
My idea was to copy all those transcriptions from the video to a notepad file. Then strip down the time-stamps using a clever program that i created. Then i copied the contents from my notepad to the freelancing website,successfully converting the content from video to text. I also improved on the grammar and other mistakes that YouTube might have done while auto-generating the transcriptions and submitted it. In the end i got a 5 star rating!
I realized we could use the same concept while converting an audio to text file.
Getting Started with converting audio to text
This process involves 3 steps –
1) Convert your audio file to a video file and upload it to Youtube
2) Copy the transcriptions
3) Remove the time-stamps
Step 1 – Converting the audio file to a video file
To convert our audio file to video file we are going to use the website called TunestoTube.
- Go to the TubetoTunes website and click on Connect Youtube account.
- Select the YouTube account that want the video to be published on and give access to tubetotunes to post on your account.
- Then upload the audio file and any image. Also give it a decent title and description.
- Click on ‘Create Video’ which will automatically create and upload the video to the YouTube account
Step 2 – Copying the the transcripts
There are two ways of copying the transcription.
Note – The transcriptions won’t appear immediately after uploading the video. Give YouTube at least 5 minutes to auto-generate the transcriptions and then go back to the video. Then you can implement the methods given below.
Method 1 – Natural
Use this method if the length of the audio you are trying to convert to text isn’t too big. To copy the transcripts using this method first go to the respective video on YouTube. Pause it and click on the ‘More’ option. Then select the ‘Transcript’ option.
Then select all the transcript text using your mouse (select and drag) and copy it. Make a text file on your desktop called ‘myfile.txt’ and paste the transcript into the text file and save it.
Use this method if the audio file is too large.
1) Go to this subtitles downloader website and paste the link of your video and click on the download button.
Note : To get the link of your YouTube video, go to your video and copy the URL (in the top-left corner) . That is the link the subtitles downloader website is asking for.
2) The downloaded file will be a zip file. Extract the contents and inside the folder you will find 3 files. We need the file that has a name same to that of your YouTube video with a ‘.srt’ extension at the end.
3) Copy that file to the desktop and rename it to ‘myfile.txt’.
Step 3 – Remove the time-stamps
If you open the ‘myfile.txt’ file saved on your desktop. You will see that the transcript of the audio is there but along with the time-stamps, which is honestly a little annoying. So we are going to strip down those bastards and make our file pure text.
1) I have created a batch file/software that will automatically remove the time-stamps from the whole file.
Download link – Timestamp stripping software
2) Extract the file and paste it on the desktop along with ‘myfile.txt’ . Then run the ‘stampstrip.bat’ file that you have pasted.
3) This will make a new file. Open the file named as ‘newfile.txt’. It will contain the final transcript. This means you have successfully learned how to transcribe audio to text.
4) Make grammar corrections and change text according to your need.
That’s it people. Hope you found this article titled ‘Convert audio to text‘ useful. Remember you can also you this same concept to transcribe video to text. If didn’t understand any of the steps above you might want to watch this video.
If you have any doubts or you just want to thank us ( this is what keeps us going) , comment below and we will love to get back to you.
If you’re into hackstories, sign up for the hackstories email list and we’ll send you the new posts right when they come out. It’s a very unannoying list, don’t worry.