But you could do this another way that would not necessitate repairing the result.
You have to open the Network Monitor, then reload the web page.
Copy Response places the json into your system clipboard. It is just plain text. Paste it into your favorite text editor. I use Notepad++ instead of plain old Notepad for many reasons, some of which you will see here. So here's what the json looks like:

It's way too off-topic for this forum for me to explain how to install & use Notepad++, & how to use Notepad++ plugins. I refer you to this web site:
https://notepad-plus-plus.orgI don't need to deal with the dog's breakfast that is this json, which is what pretty much all jsons look like. I executed the Format JSON entry on the menu you can see in that image.

Notepad++ has a nice feature that kicks in when you highlight a string. It highlights the string you have chosen, but it also highlights all other occurrences of the same string in the file. In that image, I highlighted the first manifest URL. Notepad++ simultaneously highlighted the other 3 occurrences of it. So the first manifest URL is just repeated.

Experience has shown that you can pick either the akamized manifest or the skyfire manifest. You will find the same results in both places. I decided to use the akamized manifest. It was first. It's basically a random choice. Don't dwell on that.
Running that manifest through ffprobe gave the results you can see in attached file ffprobe.txt. It describes a single WebVTT captions track, a single audio track, & 7 video tracks of various resolutions.
I was surprised to discover that there are actually English subtitles being offered here. I was hoping they would be the lyrics to the song since I couldn't make out any of them. To my ear, they could have been in any language at all. I ran the 2 invocations of ffmpeg you can see in attached files Vimeo download (ffmpeg)mp4 Log.txt & Vimeo download (ffmpeg)vtt Log.txt
A word about attachments. I attached the files to this post in the order in which I mention them in this narrative. Google, in its infinite wisdom, has decided to place them in reverse order. This is what always happens. So annoying. So look at the file names, not their positions below.
For the sharp eyed among you, I was taking so long to write all of this that the manifest timed out. I had to reload the web page, which generated all new objects in the Network Monitor. So the manifest used in the ffmpeg invocations is not the one I'm showing above in the Notepad++ screens. That's not unusual. This sort of thing happens all the time on lots of web sites. It doesn't change this little tutorial. The URLs I'm showing won't work for anybody else anyway. Not only do they time out, as I just found out. But they also uniquely identify my web session to the server. You need to visit the web page yourself & do all of this. Do not bother copying the URLs out of the things I've posted here. They won't work for you. What will work is the process I've described here.
Unfortunately, the captions aren't really worthwhile. Here is the entire subtitle file:
----------------
WEBVTT
02:50.363 --> 02:51.493
They're coming down.
----------------
Looking again at the ffprobe of the manifest, I saw that the captions stream is labelled as being "auto-generated." The technology for machine recognition of lyrics appears to be not particularly reliable, at least not today. But if you encounter a case like this one, in which captions are offered, this process will get them for you.
In any case, here is what these new results look like:

This is the same video as I got by repairing the one VDH downloaded, & VLC confirms that. The difference is I did this in one step without the need to repair the results.
So you've got a choice:
Download it with VDH then repair it.
Download it in one step with ffmpeg. If there are captions, which VDH doesn't handle, you can get them with ffmpeg.