I have a manually transcribed text file of a audio/video file. Additionally, I used autosub to create a srt file that I can add to the video. BUT autosub transcription algorithm is missing about 50% of the words but does have the timing right. So I'm wondering if I can somehow merge the correct transcription file with the correct timing from the srt file.

Hope that makes sense.


You can use almost any text editor to merge by hand, (the SRT file should be plain text) - just avoid Notepad, Wordpad & Word as they are likely to insert strange content, or you could possibly use a python script to provide interpolation between time stamps.

If the autosub is simply missing words then you can take the time stamps from it to insert them before the matching words, the SRT format is easy to create by hand or script.

Where there a lengthy blocks of words you can split them into lines and divide the time between the actual time stamps by the number of lines to get the interval between them and increment the time stamp from the first.

  • 1
    I'll try this. My best option is to write a script that will insert the matching timecode with the text. So I don't have to do it manually. Thanks Steve! – Javier K. Sep 19 '17 at 0:38
  • @JavierK. You might also like to take a look at github.com/RafayGhafoor/Subscene-Subtitle-Grabber as it may be useful to you. – Steve Barnes Sep 20 '17 at 4:41

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.