Can Video DownloadHelper rip the subtitles?

[klingf]

Your question

I love to watch danish television "www.dr.dk". To better understand the danish language I turn on the subtitles. Is there a chance to rip the subtitles embedded in the stream with Video DownloadHelper?

Addon & CoApp version and details

Version 8.2.1.1 for Firefox 123.0.1 on MacOS 14.3, 64 Bit

[GrampaWildWilly]

VDH does not currently support captions but we've been promised that at some future unspecified date. In the meantime, it is possible -- SOMETIMES -- to get subtitles with ffmpeg. Search this forum for "Plan B" to find how I have downloaded certain items with ffmpeg when VDH could not. Some of those instances stumbled upon captions. I repeat. SOMETIMES you can find the subtitles & download them with ffmpeg. Every case is different so there is no blanket one-size-fits-all solution.

[mtomsic]

It's possible to download the subtitles separately. Just make sure to turn on the subtitles to make the file appear.

[mtomsic]

Looked at the file, and apparently it's not complete. After fast-forwarding a bit, segment_1.vtt appeared. The URLs are, however, identical except for the incremented number.
segment_0.vtt - URL
segment_1.vtt - URL

If you really want the subtitles, I would create a script that increments the number and stops when the HTTP response code ≠ 200.

[mtomsic]

Also managed to find this playlist.m3u8 which makes things easier.

[klingf]

Sorry, but where to do this. I am not a DLH user for long time and I am not used in network monitoring tools. How has this to be done??

[GrampaWildWilly]

Unfortunately, there's no shortcuts. Do what I suggested above. Search this forum with Plan B as your search key. You will find examples of what you have to do. It's not a neat, tidy GUI process. You have to be a tinkerer, a do-it-yourselfer, somebody who likes to learn new things. Good luck.

[zbyna]

Just have the same sort of problem, but solution was different:

• all subtitles obtained in one request - see: response tab
  

[GrampaWildWilly]

Yes, some sites don't use a standard approach. The type shown for the file in your image is octet-stream. On a site that adheres to standards, that file would have a type of vtt. You're not showing it in your image but what was the URL of the item you highlighted? Did it have the string vtt somewhere within it?

In general, if you can find an item that has the captions, you can get it with ffmpeg. Alternatively, you can double click the item in the Network Monitor. That will open the item in a new browser tab. It should show you the subtitles. Then you can Ctrl+a to highlight the full page, then Ctrl+Insert to copy the text to your clipboard. Text. Captions are plain text. Then go to your text editor -- Notepad or something like that, NOT a word processor -- & Shift+Insert to paste the subtitles into a file. Then save it with the file extension .vtt. The location of the file should be wherever you downloaded the associated video. The name of the file should match the name of the video. So if you downloaded a video & gave it a name of My Video.mp4, your caption file should have the name My Video.vtt in the same directory. When you play the video in VLC, it will automatically find & display the subtitles.

[zbyna]

Thanks for the response. The string VTT is at the beginning of response containment.I will post the URL when I will get to my computer.and yes I used Windows clapboard for copying subtitles from response to new text file created in notepad. But before selecting all the lines I had to scroll down to the end of response in order to have the right char set I mean the right diacritics, in case of not scrolling the diacritic was distorted at the end. I rename the file according to the video file with extension SRT.

[GrampaWildWilly]

I do see certain letters with diacriticals at the top of what's in your image. I don't understand what you mean by scrolling to the bottom to get the right character set. I might recommend Notepad++ to you. Notepad++ replaces Notepad & has a number of improvements over the basic Notepad. The one relevant improvement is that with Notepad++ you can set a file's character encoding to ANSI or UTF-8 or a number of other things on demand. For your caption file, I would choose UTF-8.

In this particular case, you should not give the file an extension of SRT. These are not SRT captions. The SRT format uses 3 lines per caption. The second & third lines are as you see them in your image. But the first line is a sequence number, starting at 1. Each caption is numbered in the SRT format. There are no sequence numbers in your image so you should call it VTT because that's what it is. Calling it SRT doesn't make it SRT. Without the sequence numbers, you're just giving the file an incorrect name.

My personal preference is VTT. VTT & SRT are identical except for the sequence numbers. The sequence numbers are simply a headache & give you no advantage. If anything, SRT is not as good as VTT because of the sequence numbers. SRT files are always larger than their equivalent in VTT because of the sequence numbers. But everybody has their preference & there's certainly nothing wrong with SRT. It's certainly a format you'll encounter frequently. However, in the case of what you're showing in your image, they are not SRT captions. They are VTT.

[GrampaWildWilly]

Curiously, the first line of a subtitle file in both VTT & SRT format is WEBVTT. This is kind of confusing. SRT files begin with WEBVTT same as VTT files. I don't understand that but that's the way it is.

[zbyna]

This is the URL for subtitles:

For editing text files, I have been happily using gvim for years. But diacritics was distorted in both gvim and notepad, but only for parts which were hidden in Firefox Response panel. I tried to scroll down to the end of the text of Response panel before copying all subtitles and character set was OK then. It may be some issue during copy from Response panel to system clipboard.

[GrampaWildWilly]

OK. The URL for the subtitles ends in .vtt which would normally indicate they are VTT subtitles. And in this case, that is indeed what they are. VTT, not SRT. What is non-standard is that the type of the file shown under the Type column says octet-stream when it should say vtt.

When it comes to special characters, I think you will find that Notepad++ does a very good job of handling them. You would do well to hunt that down & try it out. It has an option for easily changing the character encoding of text files from ANSI to UTF-8. UTF-8 is probably what you would need with these subtitles. That's Notepad++. The 2 plus signs are part of the name of the application.

To copy the Response from the Network Monitor, you should pop up a context menu on the item in the left side of the display. There is a Copy Response selection on the cascade submenu below Copy Value. I think you'll find that more convenient than any other way of getting the data into your clipboard. But I believe any problems you're having with the special characters are due to the text editor you are using. Notepad is probably woefully inadequate. I've never heard of gvim so I can't comment on that. But I've been using Notepad++ for several years & it handles special characters quite easily.

[zbyna]

This is gvim(vim Version), simply highly not probable that vím does not handle Unicode well. I blame the windows thing 🙂 Anyway scrolling down the box solves this case.
Thanks for response.

[RomainLG]

Hi there, I'm trying to rip subtitles from https://gem.cbc.ca/saint-pierre

I don't see any vtt, or anything related to subtitles.
Only AAC and mp2t, Response tab on those show a bunch of characters.

Would you happen to know how to do ?

[GrampaWildWilly]

Yes. ASS. If I recall correctly, that stands for Advanced SubStation Alpha. I don't really know what that is but that's what I've read. For converting MP4 to MKV, that's easily done with ffmpeg. Just specify your MP4 on the -i & name your target file .MKV. One of the few things that is simple with ffmpeg. If you have .ass subtitles, I would expect Subtitle Edit would instantly convert those to srt or vtt. I predict you wouldn't have to manually fiddle anything to do with extraneous \r & \h with SE, but I can't claim to have actually done anything like that. I think SRT is simply older than VTT. VTT came along later & got rid of the entirely useless complication of the sequence numbers. It's perhaps a rare case of progress actually being an improvement. Most of the subtitles I've encountered have been VTT. And when they aren't VTT, I convert them to VTT with SE if I'm going to keep them. The only place I've seen talk of ASS subtitles has been in the SE forum here on GitHub. I don't have a sense of how widespread they are in the wild, but they do seem to be one of the standard formats.

Meanwhile, you haven't mentioned anything about subtitles on the show itself. Any progress with that?

[GrampaWildWilly]

Oh. I'm idly wondering if SE could have extracted your closed captions from your MKV.

[RomainLG]

When I said I succeeded I meant on the full show ! Hence my question is solved.
I'll try extracting the subs from the mkv with SE, maybe I can shave a step !

There's no doubt ffmpeg can convert the mp4 to mkv but that wouldn't rearly be different than having handbrake do the same

[GrampaWildWilly]

OK. I was just wondering whether the full show in fact had closed captions, same as the trailer. As for other tools, my suspicion is that a lot of them have ffmpeg hiding inside of them. VDH does, SE does. I'm sure there's others.

[GrampaWildWilly]

Indeed.

HandBrake/HandBrake#6452

It appears ffmpeg is lurking inside Handbrake.