well i've decided i am just gonna manually paste all my created html files of the j2 con panel youtube transcripts into blogger because it's silly how much time i put into it and it only being available to me locally. and realistically, i'm very fast and efficient at data entry stuff, so me just pasting 500 some odd files and doing a few bits of typing with each one won't take me very long
and also gotta refresh my memory on how i made them to begin with so i can grab the latest panels since i did the big bulk go
edited to add, i had to move it to wordpress so i could get exact phrase searching womp womp
Anya is live and ready to show you everything. Watch her strip, dance, and perform exclusive shows just for you. Interact in real-time and make your fantasies come true.
✓ Live Streaming✓ Interactive Chat✓ Private Shows✓ HD Quality✓ Free Actions
Free to watch • No registration required • HD streaming
well, while i've been twiddling my thumbs waiting for google to index the transcript files, i could have been fixing all the whacky ass characters that were in the filenames due to stuff like pipes and quotes in the youtube titles. i'd removed all (i thought) of the slashes but i didn't realize how many other things had been translated into special chars until i made a page with direct links to all the transcripts by doing something quick in libre-excel instead of sed or whatever and it didn't translate them. woops
me realizing that blogger's search doesn't do exact phrases. why do all roads lead to the google programmable search / search console (and google refusing to index my ugly ass transcript files)
could not for the life of me figure out how i made this links to all transcripts page that i was using as a way to easily get to the html pages to put in blogger (hit the post limit on the day so i'm doing other tasks now) and have to go digging through my old posts talking about this to find out i did it in libre-excel??!?! lol
past nic could you whisper in my ear how the fuck you did this plz
edited to add, i found the ods file and figured it out from there. it was less complicated than my addled brain was thinking, but also more complicated in a way because i couldn't just paste it into the thing i had swap single for double quotes and scp stuff around like a heathen
my memory is *trash* but one of the few things it does retain well is computer stuff and man, it's reassuring to know i remember enough to like, transfer files over to the linux server and get their permissions sorted for web. and if i can't remember exactly, i know the right thing to search for. still not sure this is the way forward, but i do have a few files to test with and a little google programmable search thing pointed at it. figured i'd stop making my head hurt trying to consider all the things i'd *like* to do (and need to do?) and just see if i can get the very basic part working first and then go from there
Anya is live and ready to show you everything. Watch her strip, dance, and perform exclusive shows just for you. Interact in real-time and make your fantasies come true.
✓ Live Streaming✓ Interactive Chat✓ Private Shows✓ HD Quality✓ Free Actions
Free to watch • No registration required • HD streaming
all right so i had this sudden need to offer my searchable jared and jensen panel transcript files to someone out here on the internet, but right now it's all just living as 500+ .srt files on my computer that i use grepWin to search the text of with.
i am wracking my brain trying to figure out a way i could possibly put these things on the web somewhere, they're essentially just text files with timestamps, that would be searchable and have links to the source videos. without having to do a lot of manual work and also without having to build something up from scratch.
ruled out some options (was thinking about a free wiki type thing, blogger, google drive). but i've got a domain that's already got some web facing stuff on a little linux shell, and thinking about if i have the files there then i could maybe leverage something like that google programmable search engine like the person that made the search superwiki transcripts only
that would at least get them public and searchable relatively easily? hm. i'd like to at least provide a link to the video too. will continue thinking on it
downloading youtube subtitles/captions en masse (or videos, with diff options) & searching them
i wanted to take a playlist and download all of the autogenerated captions so i could then search them. here's the solution i've come up with. tried to cover even the basics here and not assuming you have any particular experience with installing/using stuff in command line
all these instructions will be for windows (11), the wiki has instructions for other OSes
install yt-dlp
this can be used to download youtube and thousands of other sites' videos with a high level of customization, and fortunately for me also allows the option of downloading the subtitles without the video. i didn't really want to download 600+ videos for my project!
https://github.com/yt-dlp/yt-dlp -> download the yt-dlp.exe (see for other OSes)
put it in an executable path, i just added a new one to make things easier, see under the cut for detailed instructions
install ffmpeg
this is needed for a variety of things, but specifically for my case, i need it to convert the .vtt file that youtube gives us (which is very cluttered to read because it has formatting info in it too) to .srt which is text and timing only
https://www.ffmpeg.org/ -> same deal as yt-dlp, ditto the executable path - i downloaded this one: https://www.gyan.dev/ffmpeg/builds -> ffmpeg-release-essentials.zip
install srt_fix (for cleaning up autogenerated captions)
to fix the doubling of text caused by how youtube creates the autogenerated captions:
before fix -> after
https://github.com/bindestriche/srt_fix -> create a folder in your path with the .exe's called 'yt-dlp-plugins' and copy the folder with srt_fix in it there (so the final path of the srt_fix.py for me is: C:\Users\ns\bin\yt-dlp-plugins\srt_fix-1.0.1\yt_dlp_plugins\postprocessor)
there are other methods of installation on that page as well as a link to the general yt-dlp plugin installation documentation
install grepWin
if you want to search the text of the files quickly (which was the whole point of this project for me 🤪) grepWin is a good option, made by the person who did tortoisesvn. at some point i might look into putting these on The Internet somewhere that would be searchable and link to the original videos/timestamps, but right now, i just wanted a quick way to search them.
https://github.com/stefankueng/grepWin - click on the release on the right
documentation with direct link to current release: https://tools.stefankueng.com/grepWin.html
now we're ready to go!
open powershell (start->windows powershell) and navigate to where you want the files to be placed, for example: (ps: press tab after writing a couple letters and it'll fill out the dir name automatically)
paste your command (using one of the panel videos for example):
replace with the video or playlist you're interested in. recommend starting with a video first :) if it seems to be bombing out repeatedly, stop with ctrl+c
the flags are needed as follows:
-t sleep - if you don't use this, youtube will stop the download after 300 videos and make you wait an hour, this adds some delays to get around that
--write-auto-sub - this is to grab the autogenerated captions as opposed to the uploaded/created versions
--convert-subs "srt" - to use ffmepg to convert the vtt subs into an srt file
--skip-download - to not download the actual video, we only want the subtitles
--use-postprocessor srt_fix:when=before_dl - this is calling the srt_fix specifically when using the skip-download flag
when it's working correctly, it'll look something like this and the folder will contain two files per video, the original srt and the -fixed srt with the non-duplicated text
the file it creates is named (by default, this can be changed, see the docs) video title [youtube video id] - aka plop that string on the end of this url and it'll pull up the source video https://www.youtube.com/watch?v=
and lastly, searching! obviously this is where things kind of break down, because common words are more likely to be understood and captioned correctly; but proper names, sketchy audio quality, mumbling, etc is going to miss a lot. just for kicks, i thought i'd see if it ever got mark of cain:
detailed instructions on adding a new entry to the path environment variable below - just be careful with this
open system properties - search for that in the start menu or enter environmental - it'll pull it up
in the system properties window -> click environment variables
double-click the user variable Path
click New and put the path. i made a new folder at my profile folder level (ns) called 'bin' for this:
well i've got 200/618 (now that i got the panels after june 2025 as well) transcripts up. and an interesting thing too doing this manually is i can find out when the source video is gone now and either track down an archive.org copy, gayle's copy archived or (hopefully) an alternative on yt
it felt kind of redundant having the embedded youtube video and the regular link to it, but having the link easily accessible also means if you run into something that's no longer around on youtube you could also just pop that url into the archive.org search
