Need help with this private embeded vimeo video for downloading

[Supernaut]

Here is the file attached which is from the embedded private vimeo video on a private members only access site. I followed all the directions but still can't figure it out. I was able to download it from the dash and use ffmpeg to make frame rate normal to 25 (and it worked) but it takes like 30 min per video that way.

Can someone show me step by step how to grab the vimeo link from this file so I can download it? I looked at so many forums on here but none of them explain it like how I need to tackle this file.

Please let me know or if you need more info.

Thank you.

[Wild Willy]

In your file attachment there are 7 occurrences of the string m3u8. A file of type m3u8
is an HLS manifest, most likely an HLS master manifest. Although there are 7 occurrences
of m3u8, they are all within just 2 distinct URLs. Since I can't look at the manifests
myself, what with them being for a private Vimeo video, you'll have to do it. You will
have to get ffmpeg & learn how to use it. I have posted a tutorial on ffmpeg in this
forum. You can find it by starting here:

https://groups.google.com/g/video-downloadhelper-q-and-a/c/BzPLK2YyL-s

Look for "cannot download" within the text of that web page. That will give you a link
to the ffmpeg tutorial. As I developed that tutorial, I learned things. Some stuff I
originally thought was important, but as I learned more I realized it wasn't as
important. I also learned how to do certain things more simply as I went along. So
you'll want to read the entire tutorial before you jump in & try to do any downloading
with ffmpeg.

I am encouraged that you figured out how to "respeed" your video. It seems you've read &
understood this thread:

https://groups.google.com/g/video-downloadhelper-q-and-a/c/sNfTCMYfiTU

It's always reassuring to know that somebody can follow what we had to say there.
However, half an hour to respeed a video seems a bit excessive. Either your hardware is
way too slow or your videos are a few hours long. But if you can learn how to use
ffmpeg, respeeding should not be necessary.

[Wild Willy]

I should rephrase some of what I said a bit. You do already have ffmpeg & you have used
it to respeed your videos. But ffmpeg can do a lot of things. I was suggesting you
learn how to use it to download videos. That's the focus of the tutorial I mentioned.

[Supernaut]

Thank you.

I read through your tutorial on how to find the manifest and then find what you need to put into ffmpeg to download the video. But it looks like I can not find the manifest when i type m3u8 into the search under Network and All like you said. I also refreshed the page like you said. Here is a pic of what I see. Does this mean I won't be able to download the video like how you mentioned that some sites will not provide the manifest to counter people from downloading their videos?

[Supernaut]

I guess using VDH and ffmpeg to change the frame rate manually will be the only way. I have no computer science background, I am actually a nursing student lol. The most I have done with code is learning how to use adobe flash when that was a thing to make a soundboard like 6 years ago.

[Wild Willy]

As long as you don't flash your patients, you should be fine.

Vimeo is weird. They don't put their manifests directly into the Network Monitor. But
in your first post, you attached a file that DOES contain the URLs of 2 HLS manifests,
the first one 3 times, the second 4 times. However you found that file is the key to
unlocking success here. In the second thread I cited above, the one with the title about
too fast video, we explain how to find that weird Vimeo file that contains the manifest
URLs. I have experimented a bit with the Vimeo examples we were presented with in order
to come up with that discussion. My experiments showed that the 2 distinct manifests in
that weird file were equivalent. You could use either one to get the same content. Once
you've got the manifest URL from there, it's a piece of cake to download the file with
ffmpeg. The rest of the ffmpeg tutorial will tell you how to proceed. In normal cases,
you can find the manifest in the Network Monitor. In the case of Vimeo, you have to look
for the weird file in the Network Monitor, & the URL of the manifest is inside that file.
Otherwise, the instructions are the same in both cases.

[Wild Willy]

There is one tricky bit I should point out. Let's take the example of the first manifest
URL in your file you attached to your first post:

\"https://174vod-adaptive.akamaized.net/exp=1677467958~acl=%2Fe44d4cef-3a16-4d4b-91cc-bb0adfd40005%2F%2A~hmac=8a210951ae97a0cb2bde2ea06400c23aa7ec014c6733ad017d2b16f6ccb0a47c/e44d4cef-3a16-4d4b-91cc-bb0adfd40005/sep/video/3d53e023,a1b08b14,c3e14cbb,cccc5e55,d76428d5/audio/92adc3d6/subtitles/74789014-English%20%28auto-generated%29-en-x-autogen-cc/master.m3u8?external-subs=1\\u0026query_string_ranges=1\\u0026subcache=1\\u0026subtoken=e50dda4c13494ff03dd8ab0272df94bc5202ef7945ee786abcb53f9d510dfb86\"

If you can keep your eyes from spinning out of control, you will see master.m3u8 inside
there. The m3u8 tells us this is an HLS manifest. The further appearance of master.m3u8
means it is almost certainly a master manifest.

First off, whatever language the weird file is written in, it uses \ as an escape
character. That means that, for example, the leading \" really means just ". Similarly,
the trailing \" really means just ". But " isn't part of the URL anyway. It's just that
in order to figure out the beginning & end of any URLs in that weird file, you need to
look for the \" pair surrounding the URL.

You'll see a few instances of \\u0026 within the URL. Again, \ is an escape character.
The real string is \u0026. But even that is kind of smoke meant to confuse us. \u0026
is really just &. I had to Google that to learn that little tidbit. Try it yourself.
Do a web search using \u0026 as your search key. So your URL is really this:

https://174vod-adaptive.akamaized.net/exp=1677467958~acl=%2Fe44d4cef-3a16-4d4b-91cc-bb0adfd40005%2F%2A~hmac=8a210951ae97a0cb2bde2ea06400c23aa7ec014c6733ad017d2b16f6ccb0a47c/e44d4cef-3a16-4d4b-91cc-bb0adfd40005/sep/video/3d53e023,a1b08b14,c3e14cbb,cccc5e55,d76428d5/audio/92adc3d6/subtitles/74789014-English%20%28auto-generated%29-en-x-autogen-cc/master.m3u8?external-subs=1&query_string_ranges=1&subcache=1&subtoken=e50dda4c13494ff03dd8ab0272df94bc5202ef7945ee786abcb53f9d510dfb86

If you decide to use a command script, as I strongly urge in that tutorial, you will also
need to double all the % signs, a topic I touch on in the tutorial. So your URL would
become this:

https://174vod-adaptive.akamaized.net/exp=1677467958~acl=%%2Fe44d4cef-3a16-4d4b-91cc-bb0adfd40005%%2F%%2A~hmac=8a210951ae97a0cb2bde2ea06400c23aa7ec014c6733ad017d2b16f6ccb0a47c/e44d4cef-3a16-4d4b-91cc-bb0adfd40005/sep/video/3d53e023,a1b08b14,c3e14cbb,cccc5e55,d76428d5/audio/92adc3d6/subtitles/74789014-English%%20%%28auto-generated%%29-en-x-autogen-cc/master.m3u8?external-subs=1&query_string_ranges=1&subcache=1&subtoken=e50dda4c13494ff03dd8ab0272df94bc5202ef7945ee786abcb53f9d510dfb86

But that's only if you use a .bat file. If you handle this some other way, you wouldn't
double the % signs.

The URLs you are likely to encounter almost always contain strings, sometimes rather long
strings, of gibberish letters & digits. These strings uniquely identify your online
session. They connect your browser session to the web server. They may even somehow
contain your user ID & password in some sort of encrypted form. It is also common for
these gibberish strings to expire. The URL I'm quoting above from your file attachment
no doubt pointed to the manifest we want. But it no longer does. You have to visit the
page again to get the web server to generate a new manifest with new gibberish in its
URL. Then you have to give that URL to ffmpeg within a reasonably short period of time,
usually within 20 minutes or so. Otherwise, you have to reload the page & start over.

I also see mention in that URL of "subtitles" & "auto-generated." I suspect there are
captions associated with your video. We can figure that out once you run your manifest
URL through ffprobe. It will tell us where the captions are, if they exist. On the
other hand, if they are automatically machine generated, I wouldn't be surprised if they
turn out to be pretty worthless. But we can't know that ahead of time. We have to get
them & see how well they match up with the video content. The captions might be closed
captions. I do see cc in that URL. If they are closed captions, you will just get them
as part of downloading the video. Closed captions are typically embedded within the
video track & you just get them automatically when you download the item. Only if
ffprobe tells us there are external captions would we need to download those in their own
invocation of ffmpeg.

Are we nearing enlighentment yet?

[Supernaut]

I appreciate the effort you put into telling me how to do this. Let me try it following your tutorial and see if I can get it. Thank you.

[Supernaut]

When I press enter for this cmd prompt for ffmpeg, i get a error "403 forbidden (access denied)" I attached a photo. I attached the new file I just got too, let me know if I did it correctly.

Here is the command i ran.

ffmpeg -protocol_whitelist file,crypto,data,http,https,tls,tcp -hwaccel auto -i "https://174vod-adaptive.akamaized.net/exp=1677574583~acl=%%2Fe44d4cef-3a16-4d4b-91cc-bb0adfd40005%%2F%%2A~hmac=bc774e5fc4e1bfb939fd15b8cacb7744009815960c65d6d89ac226b15f0c01ff/e44d4cef-3a16-4d4b-91cc-bb0adfd40005/sep/video/3d53e023,a1b08b14,c3e14cbb,cccc5e55,d76428d5/audio/92adc3d6/subtitles/74789014-English%%20%%28auto-generated%%29-en-x-autogen-cc/master.m3u8?external-subs=1&query_string_ranges=1&subcache=1&subtoken=e50dda4c13494ff03dd8ab0272df94bc5202ef7945ee786abcb53f9d510dfb86" -codec: copy "C:\Users\Pizza\Downloads\Video Download Helper\output3.mp4"

[mjs]

That is unfortunate, the server for the master manifest at akamaized . net has forbidden access to the manifest. Meaning it's off limits to

everyone. So you've hit a wall on that one , you'll have try a different method to get the video.

[Supernaut]

All good. I appreciate your guys help. I will use VDH to download the videos and then use ffmpeg to change the framerate to 25. That seems to be working except on 2.5 hr long video had the audio and video off after using ffmpeg to change the framerate. Better than nothing.

[mjs]

Hope that you learned how to find correct frame rate before using ffmpeg. There can be no guessing on it otherwise the video and audio won't

be in sync. I learned that myself after the ffmpeg fix was posted on the forum.

[Wild Willy]

Well . . . maybe.

I see that you doubled all the % signs. But it looks like you just executed the ffmpeg
command in a command window. You need the double % signs ONLY when you execute ffmpeg
from a .bat file. You didn't use a .bat file. You just entered the ffmpeg command into
the command window. Or did you copy/paste it from somewhere else? In any case, the way
you are doing it means you must NOT double the % signs.

Also, it looks like either one of 2 things:

- Whatever directory contains ffmpeg is in your system PATH.
- You have put ffmpeg into directory C:\Users\Pizza\Downloads\Video Download Helper.

I just hope you are NOT using the ffmpeg that is distributed with VDH. That will not
work. You need to get ffmpeg as I explain in the tutorial. Then where you put it is up
to you. I see in your image of the command window that ffmpeg is reporting a date of
February 22, 2023 & it's a gyan.dev build. So yes, you did correctly get a real ffmpeg.
I'm just curious where you put it on your system.

You said at the outset that this is a private Vimeo video. That means to me that you are
logging on to some web site you have not named & giving your user ID & password. I
suspect that you have to supply your user ID & password to ffmpeg. I cover that in the
tutorial. Just search for "password" within the text of the tutorial.

So I'm suspecting 2 things might help here:

1. Don't double the % signs.
2. Include your user ID & password in your invocation of ffmpeg.

Might. It's possible it won't work, as mjs said. But it's worth a try. I feel like
you're very close to success.

[Wild Willy]

If you do end up using your user ID & password, don't post another screenshot of your
command window like you did above. It will show you user ID & password to the world.
You don't want to be doing that. So if you do post something, make sure you obscure your
user ID & password in the image before you post it. Same thing if you quote your ffmpeg
command. Replace your user ID & password with something, anything else.

[Wild Willy]

It occurs to me that you should do this in phases. Phase 1 would be to revert the %
signs to singles, then see if it works. I would do it twice, once the way you did it,
then reload the page & go through the whole process again. As I mentioned before, these
things have a tendency to expire. I have encountered a site on which the manifests
expired every 10 seconds or less. But that was a livestream & that's not what you're
dealing with here. Still, just to be sure, I would do a reload & retry before concluding
this does not work. Going with single % signs is definitely a correction. Whether it
works is subject to testing.

Phase 2 would be to add the user ID & password. I think there is a possibility this is
not going to be necessary. Also as I mentioned before, the gibberish strings can
sometimes include encrypted authentication information. So there's a possibility you
won't need to add your user ID & password. Only if phase 1 fails twice, then include the
user ID & password. And as before, do it at least twice, with a reload in between, same
as above. It could be that your user ID & password on the original site is something
akamized doesn't know about. In my estimation, that is quite likely. You're a member of
your site, not akamized. It would be a security risk for akamized to carry a duplicate
user data base with passwords. This would say to me that the gibberish in the URLs
indeed does include authentication information. However, you never know. So if phase 1
comes up empty, try phase 2.

[Supernaut]

Holy crap guys... it worked!!! I took out the double %% and it just downloaded in 5 min or so! Here is the command I ran for a new video I just grabbed the master manifest from:

ffmpeg -protocol_whitelist file,crypto,data,http,https,tls,tcp -hwaccel auto -i "https://47vod-adaptive.akamaized.net/exp=1677651992~acl=%2Fdc72a621-c103-4afa-a2e8-6502e9d1b0c6%2F%2A~hmac=4fdf67f8bc8d659606c20a89ab90ae74bbc31eb153236f99318e51e40c63e3a9/dc72a621-c103-4afa-a2e8-6502e9d1b0c6/sep/video/240ccd60,71d24828,9d0cbc07,da72326c,eafd6a21,faf4cf39/audio/dc23b8aa/subtitles/76055403-English%20%28auto-generated%29-en-x-autogen-cc/master.m3u8?external-subs=1&query_string_ranges=1&subcache=1&subtoken=3bc6852c6bea0c75eefa4df72c21b5f8ae74a64db9c588cbaad839e1f3408f54" -codec: copy "C:\Users\Pizza\Downloads\Video Download Helper\output3.mp4"

Thank you guys so much! I can download everything like nothing now! :D

[Supernaut]

Looks like I ran into another problem. When I go to the private websites video list and try to get the weird file with the gibberish by putting "player.vimeo.com" into inspect element, no matter what video I click on, the weird gibberish file will only be for the latest video. Meaning I can only find that gibberish file and not previous video gibberish files. Please let me know how to get around this. I tried pressing refresh on it multiple times and and it does not work. I attached a pic. It shows more than 2 gibberish files to download because I clicked on multiple videos to show you what pops up each time I click on a different video. Is there a workaround? Please let me know.

[Supernaut]

I figured it out. I just used another gibberish file without using "player.vimeo.com" in the filter. Thanks so much Wild Willy! I literally could not do any of it without your help because I don't know anything about computer science! lol

[Wild Willy]

I wouldn't say "nothing." You are getting to be an expert in ffmpeg. And I congratulate
your improvising to solve your problem. It's a good thing you figured it out because I
don't know that I could have advised you without getting my own hands on the site. You
may be more of a computer geek than you give yourself credit for.

In any case, I'm very happy our instructions, both mjs's & mine, resulted in your success.

[mjs]

Great that you succeeded but hang on a minute, isn't that a json url and you were successful to get a video with it

Those master json links have never worked when I tried them in my testing. Here is another I gave to ffprobe just now :

 Syntax error near offset 12.
https://37vod-adaptive.akamaized.net/exp=1677648058~acl=%2F3cbb353c-eb42-446c-a738-aabbf13c6a0c%2F%2A~hmac=a32f67698c6533f0d0b42f8b4d1524761ce95f92b6373fe9acc5d0360c92f79c/3cbb353c-eb42-446c-a738-aabbf13c6a0c/sep/video/d309eae1,0ae188d2,af7a3b51,b5899eec,67eeef13/master.json?base64_init=1: Invalid data found when processing input

[Supernaut]

You need to have that sub token sub cache in there. You can only get that when you download the gibberish file. I am running into a ouput error suddenly as well. it came up out of nowhere while exporting my 6th or so video. Here is a pic and the command

ffmpeg -protocol_whitelist file,crypto,data,http,https,tls,tcp -hwaccel auto -i "https://186vod-adaptive.akamaized.net/exp=1677661462~acl=%2Ffbac9b9b-15b3-4dfa-84be-0979fb2cadb9%2F%2A~hmac=b4fce455c1ca9b6819350d4c41ba0a463c5a9930c858e3b0d9e74b658628e44d/fbac9b9b-15b3-4dfa-84be-0979fb2cadb9/sep/video/ef143119,2e3e33cb,c0a30425,e95a9e15,eab6c206/audio/243dff4e,a2643f35,b1d47e84/subtitles/64713779-English%20%28auto-generated%29-en-x-autogen-cc/master.m3u8?subcache=1&external-subs=1&subtoken=8ac793e57eff395c6153a7eceea8e33b6fa6aad96bcf062a5437c10c5b8d11a4&query_string_ranges=1\" -codec: copy "W:\Downloads\C Trading\1 5 23.mp4"

[Supernaut]

Nvm. Looks like I had a typo in my command somewhere and I just reset it. All good here.

[mjs]

That's a master manifest , but in your attached image you highlighted a master.json url and those do not work in ffmpeg.

[Supernaut]

I know. I only highlighted it to show where I grabbed the master manifest from.

[Wild Willy]

Your sample ffmpeg command shows the -i URL ending with \. \ is not a valid character in
a URL. The problem is that json uses escape characters, like I mentioned above. You
forgot to get rid of both characters of the \" that terminate the URL. I know about this
error because I made it several times before I got wise to the issue.

[Wild Willy]

I was able to look at the master manifest whose URL you posted in the most recent test.
I can't run it through ffprobe (but you could) because it's got partial URLs in it,
strings of ../../../../../../ at the front of each URL. That's not surprising, given
that you are looking at a private video. The structure of this manifest appears to be
that 5 video tracks of various resolutions share a single audio track. You would want to
give the URL of the master manifest to ffprobe. Even if you gave me the URL for the
master manifest, ffprobe wouldn't work for me because it's a private video. So you would
have to do it.

There are also captions in there. I don't know if you're interest in those. The line
for the captions is this:

#EXT-X-MEDIA:TYPE=SUBTITLES,GROUP-ID="subs",NAME="English
(auto-generated)",LANGUAGE="en-x-autogen",URI="../../../../../../subtitles/64713779/playlist.m3u8?d=4036.200000&subtoken=ebf195722803b55cf6d8f24c098625371bd9140571e16d219e865e3319ef05ed",CHARACTERISTICS="public.accessibility.transcribes-spoken-dialog",AUTOSELECT=YES,DEFAULT=NO

Line. Singular. Google breaks it & line wraps it to fit here, but it's just one line.
If the various bits of that are to be believed, it's English captions & they're machine
generated. Each of the 5 video tracks points to this captions descriptor. Since the URL
in that statement shows .m3u8, that means that the captions are available as a stream,
not as a simple file. But that's no big deal. The video & audio are all available as
streams, too, not simple files, & you're getting those no problem.

Out of idle curiosity I would like to know how accurate the captions are. You might also
be interested in getting them just as an exercise. You never know when some day you
might actually want to get the captions for some item. If you post a complete ffprobe
report from one of your master manifests, I could tell you how to get the captions.

[Supernaut]

to run ffprobe, do i just run the same command into it like ffmpeg? I don't need the captions btw, because these videos i downloaded have been auto generated subs and not useful. I downloaded about 51 videos total today after learning how to do this

[Wild Willy]

The syntax of ffprobe is this:

ffprobe  -protocol_whitelist file,crypto,data,http,https,tls,tcp "URL of master manifest" 2>"Q:\To Watch\ffprobe.txt"

The bit of 2>something captures the output of the command in a file.  You can code that or omit it.  I say code it because it's much easier to attach a text file to a post here than attach or embed an image.  I think that will also make it easier for me to explain how to get the captions.

[Supernaut]

Sure, here is the ffprobe file.

Here is the command I ran:

ffprobe -protocol_whitelist file,crypto,data,http,https,tls,tcp "https://142vod-adaptive.akamaized.net/exp=1677694939~acl=%2F522f41d9-9709-495a-99b9-73af6d430da5%2F%2A~hmac=0fabed4549c5ae8d3494f9252a6c5579c13623291e44536beba7202f9456330f/522f41d9-9709-495a-99b9-73af6d430da5/sep/video/bb4f5fb8,6bf7eda6,85e0dd17,9254aaa2/audio/4191f524,5bb397d5,a7037486/subtitles/19922664-English%20%28auto-generated%29-en-x-autogenerated-cc/master.m3u8?subcache=1&external-subs=1&subtoken=dc435652b39bd2e170ca2acfd5ef801b6002c27baadd2990f6a61ce940f3c406&query_string_ranges=1" 2>"W:\Downloads\ffprobe.txt"

[Wild Willy]

OK.  I can work with this.

Look for this line near the top of the ffprobe report:

[hls @ 00000207b99f0280] Can't support the subtitle(uri: ../../../../../../subtitles/19922664/playlist.m3u8?d=9712.333333&subtoken=2538b0d91105ad33a764ddb3c6f376b6b0b0ee451faca4485125d01818222dd5)

I don't know why it generates this error message.  Ffmpeg is perfectly capable of downloading captions.  Even ffprobe can generate a report on a captions file.  In any case, this gives us the URL of the captions stream manifest.  Problem is it starts with a bunch of ../ things instead of https://something.  This is an abbreviation of the real URL.  But we can get the part of the URL represented by the ../ abbreviations from the rest of the report.  There is a common part of every URL you can see in the remainder of the report:

I'm using Notepad++, not Notepad.  One of its many useful improvements over Notepad is this highlighting thing it does.  I highlight one occurrence of a string & Notepad++ automatically highlights every other occurrence of the same string in the file.  I could scroll it down & show the images but that would be rather wasteful.  You can take my word for it that every URL out of sight below this first screen load starts with this same string.  This is the part of the URL that is abbreviated by the ../ things in the error message.  So the full URL of the captions stream manifest is this:

https://142vod-adaptive.akamaized.net/exp=1677694939~acl=%2F522f41d9-9709-495a-99b9-73af6d430da5%2F%2A~hmac=0fabed4549c5ae8d3494f9252a6c5579c13623291e44536beba7202f9456330f/522f41d9-9709-495a-99b9-73af6d430da5/subtitles/19922664/playlist.m3u8?d=9712.333333&subtoken=2538b0d91105ad33a764ddb3c6f376b6b0b0ee451faca4485125d01818222dd5

To get the captions, do this:

ffmpeg -protocol_etc -i "https://142vod-adaptive.akamaized.net/exp=1677694939~acl=%2F522f41d9-9709-495a-99b9-73af6d430da5%2F%2A~hmac=0fabed4549c5ae8d3494f9252a6c5579c13623291e44536beba7202f9456330f/522f41d9-9709-495a-99b9-73af6d430da5/subtitles/19922664/playlist.m3u8?d=9712.333333&subtoken=2538b0d91105ad33a764ddb3c6f376b6b0b0ee451faca4485125d01818222dd5" "somedriveletter:\somedirectorytreestructure\some name.vtt"

The important bits to notice are these.  You don't code -codec: copy.  That doesn't work with captions.  It's not really relevant anyway.  The file name of the target file needs to have an extension of .vtt.  Actually, there are other types of caption files but .vtt is the most efficient.  If the captions on the server happen to be some other type, this will convert them to WEBVTT format.  So you don't have to figure out what they're doing.  Just code .vtt & you'll be safe.

Now, I must say that what I did to get the URL of the captions stream manifest does have a tiny component of guessing in it.  You need to try a URL & see if it works.  If it doesn't work, you need to prepend either one qualifier more or one less & try again.  A qualifier is what comes between the / characters in a URL.  This particular URL contains both / and %2f.  %2f means /.  This is called URL encoding.  You can read entirely more than you will ever want to know about it here:

https://www.w3schools.com/html/html_urlencode.asp

So you might need to go one / (or %2f) further or one less.  Rinse & repeat until it works.  You'll note that in the initial error message there are 6 occurrences of ../.  This means it is abbreviating 6 qualifiers.  If you look at the prepended partial URL I added, but change all the %2f occurrences into /, you get this:

https://142vod-adaptive.akamaized.net/exp=1677694939~acl=/522f41d9-9709-495a-99b9-73af6d430da5/%2A~hmac=0fabed4549c5ae8d3494f9252a6c5579c13623291e44536beba7202f9456330f/522f41d9-9709-495a-99b9-73af6d430da5

That's 6 qualifiers, if you count the leading  https: as a qualifier.  I'm not enough of an expert to say this works like this in every case.  But it gives you a good starting point.

While I have your attention, I want to point out something else about your master manifest.  Go down nearly to the bottom, where you can find this bit:

  Duration: 02:41:52.33, start: 0.700000, bitrate: 0 kb/s

That's the beginning of the most useful part of the report.  Notice there are 4 Programs.  That's what ffprobe & ffmpeg call them.  Programs.

Each one consists of 2 Streams.  That's more ffjargon.

But look closely at the audio Streams.

Audio Stream #0:0 appears in Programs 0, 2, & 3.  Nothing special there, just wanted to point it out.  The remaining Program 1 contains audio Stream #0:1.  The 4 video Streams in the 4 Programs are unique, their Stream numbers being #0:2, #0:3, #0:4, & #0:5.  I'm sure by now you don't need me to illustrate this.  You'll note that the video resolutions are, in order, 960x540, 1280x720, 426x240, & 640x360.

I mention the Stream numbers because I recommend you get in the habit of using them.  Ffmpeg says it will pick the "best" Streams for you. You have been allowing it to do this.  But there are times when it won't actually pick the best.  It defines best as the highest video resolution. But there are situations in which a master manifest has multiple Programs containing video Streams that are the highest in the manifest. That's not true here, but I have seen such cases.  In such a case, ffmpeg selects the first such Program for you.  It might not be the best one. Notice that associated with each Stream there is something labelled variant_bitrate.  This gives you an estimate of relative quality.  I have seen manifests in which the highest quality of a given resolution was not the first.  Ffmpeg picked the lower quality video by default, since it appeared first in the manifest, instead of the best one.  In addition to the resolution, you have to pay attention to the variant_bitrate value. The higher the value, the higher the quality.

So I always tell ffmpeg which Streams to get for me.  I never let it decide a default for me.  In this particular case, the highest resolution is 1280x720 & there is only one of them.  It is video Stream #0:3 in Program 1.  Its partner audio Stream is #0:1.  So if I were downloading this video, I would do it like this:

ffmpeg stuff -i "Manifest URL" -codec:copy -map 0:3 -map 0:1 "target file"

I urge you to always run ffprobe on whatever you're going to download & inspect the report.  Decide for yourself which Stream is actually the highest quality & then be sure to select that via -map parameters.  Don't leave anything to defaults.

[Wild Willy]

Forgot to mention one thing. In order to get the captions to display automatically, you
need to follow an easy naming convention. Suppose your video is named Video.mp4. Be
sure to name your captions file Video.vtt & put the captions file in the same directory
with the MP4. VLC automatically recognizes & displays captions when you follow this
convention: the files are in the same directory & have the same name but different
extensions. You can name the captions file anything at all & put it in any directory,
but then you have to take an extra step in VLC to tell it where the captions are.