I was mildly astonished to read reports that self-styled anti-racist grifters experts had frittered away most of the donations they had received and were now selfishly eliminating staff to secure themselves. How can you spend millions of dollars so quickly and produce nothing valuable to show for it? Among the accusations directed at the social justice warriors (SJWs) was of having treated the laid-off staff as indentured labour. How ironic is that? Descendants of slaves treating descendants of slaves as slaves! Anti-racism donations have accomplished the opposite. When someone accuses newborns of racist bigotry, then it is clear that their judgement has become permanently impaired. Why would anyone give them money? DIVIDE & RULE!
This set me thinking. Actually, every moment that I have, I think about racism. There is too much Whiteness. Thousands of miles away, in America. I had to act. What I can do to reduce racism now that the ‘anti-racists’ are operating with less-than-optimum staff levels?
Well, I have noticed or rather I have been perturbed by the prevalence of online videos of American White supremacists looking paler than the average Irish paleface. They are all over the Internets. They are always standing in front of white backgrounds. Some even place the camera so close to the face that you can literally count their nosehairs. As if this torture was not enough, they also use too much lighting. If you think Whiteness is not harmful, here is a documented case of Whiteness literally causing hurt. Sometimes, I do leave comments advising them not to do that. Only sometimes because it requires me to login to a Google account. I hate that. I advised Dale Roberts not to place the camera so high that light reflecting on his shiny dome does not instantly blind his viewers. Some Theytubers place the lights directly above them such that the shadow of their nose gives them an undeniable Hitler mustache. Wittingly or unwittingly, White people are so triggering!
Look at this lady. Can she be any more white? She is so white Dracula would turn pale on seeing her.
She is the same colour as the walls. She deliberately chose to be born with that shade of white to perpetuate Whiteness. She is the inverse of those optical illusion Calvin Klien advertisements of the 90s. The clothing would blend with the background and you could only see the exposed parts of the bodies of the models. It was so outrageous that men would stand in dark alleyways and sell illegal prints of the ads.
I took a screengrab from her video and then opened it in GIMP. I wanted to see how I can adjust the colours to make her less racist. Any adjustment to the brightness levels would require a subsequent adjustment to the contrast so I avoided that. I adjusted the saturation level until I could discern her features as distinct from the wall. She is now a person of some colour.
Dangers of low-volume videos
Another problem with Theytubers or Rumblers (not limited to Whites) is that they record the video using their phones or a laptop. This may be okay but they need to make sure not to shake or bump the recording device so as not to introduce high-volume spikes into the audio. Spikes will make the video play with low volume, as they set the bar as to how high the volume can go during playback. Ordinary volume normalization efforts will not make much headway because of these spikes. Normalization algorithms try to increase the volume without making changes to the original waveform. The spikes have already reached the maximum so normalization cannot go further, that is, without clipping the waveform.
When online video publishers (that is what Rumblers are) insert intro or outro segments, they may again unwittingly create videos that play with low volume. I have written about this problem in this month's Open Source For You magazine. In that, I have provided a Caja Action Configuration script that I use play downloaded videos with ffplay.
sFile="$*"
ffplay -hide_banner -autoexit \
-window_title $(basename ${sFile}) \
-f lavfi "movie=filename=${sFile}:s=dv+da[out1][out2];
[out2]dynaudnorm=gausssize=3"ffplay dynamically compresses the audio in realtime when it plays the video. This way, I do not have to fiddle with volume control for every new video file.
Low-volume videos by themselves are not dangerous. The danger arises when you play the next video. You may have manually increased the volume levels of your TV or speakers for the previous video and the next video (even if it has normal volume levels) will then be played at increased level that it will damage the ears and/or audio equipment.
One thing that such content creators (that is what online video publishers call themselves) can do is dynamically compress or normalize the audio before uploading. Ordinary normalization can become useless because it has to consider the whole file. Dynamic compression/normalization may be more effective as it considers smaller chunks for normalization/compression.
Do you want to reduce racism and high-volume spikes before uploading?
I have written a book on FFmpeg. How would I use FFmpeg to fix the two aforementioned problems? I would use this script.
sFile="$*"
sFilename="${sFile%.*}"
sFileExtension="${sFile##*.}"
sOutputFile="${sFilename}-filtered.${sFileExtension}"
ffmpeg -i "${sFile}" \
-filter_complex \
"[0:v:0]eq=saturation=1.6[v];
[0:a:0]dynaudnorm=gausssize=3[a]" \
-map '[v]' -map '[a]' \
"${sOutputFile}"I would run this script on a video like this:
bash anti-racist-n-compression-filters.txt \
Page-Turning-Nonfiction-The-Art-of-Reader-Avatar-Driven-Books.mp4This command would create a file named Page-Turning-Nonfiction-The-Art-of-Reader-Avatar-Driven-Books-filtered.mp4 and I would upload that filtered video.
What does the script do? It applies a colour saturation filter a.k.a. anti-racist filter (eq FFmpeg filter) to everything in the video. And, it also boosts the low-volume segments in the audio stream (using the dynaudnorm FFmpeg filter) so viewers can listen to the video without having to adjust the volume.
In this video, I have placed the original racist video on the left and the filtered anti-racist video on the right. The original audio will play on the left speaker and the compressed audio will play on the right speaker. You should adjust the balance between the left and right speaker using the the volume control utility on your device. When you move the audio balance to the left, you will realize that her original audio is very low. When you move the balance to the right, you will find that audio has been compressed and plays at a pleasantly moderate volume.
I had set the saturation level to 1.6 in this video. For another video of hers, I had to set it to 2.6. (The float range for saturation filter option is from 0.0 to 3.0.) You have to experiment with the values to obtain right amount of anti-racism.
Also note that dynamic audio compression/normalization should be applied only for videos that have a person or a group of persons just talking. It should not be applied to music. Music intentionally has highs and lows, and should not be dynamically compressed or normalized.
License
As the world's preeminent racist, I hereby authorize White supremacists to freely use this FFmpeg script and make their Theytubes and Rumbles more tolerable. Peace, y'all!
Conclusion
I think I helped. Specifically, I helped reduce competition racism. The world is a better place because of me. Everyone, hug me! No, give me millions of dollars! If not, I will also ‘peacefully’ protest and burn down some American city.
Courtesy
My thanks to Ms. Nonsense-Free Editor who unintentionally provided a good video sample.
Yes, that is an RSS link. Theytube is harmful to video creators and subscribers. If you have to use it, use channels' RSS links in an RSS reader app. The video creators will get paid for anonymous views too.
Become an FFmpeg PRO by reading my book Quick Start Guide to FFmpeg.
- MORE INFO — http://www.vsubhash.in/ffmpeg-book.html
- BUY — https://books2read.com/ffmpeg (common link for all stores)