3 ms·

I used ffmpeg and audacity with youtube-dl to download and do file and spectrograph analysis on a faked political video. That analysis showing how the audio was

I used ffmpeg and audacity with youtube-dl to download and do file and spectrograph analysis on a faked political video. That analysis showing how the audio was spliced together may have killed the story before it got legs. In an era of fakes and deepfakes, we need tools like youtube-dl more than ever.

AnonsLadder 6y ago
That's freaking awesome
brokenmachine 6y ago
Did you post an article about it? I'd be interested to read it.
lostgame 6y ago
This is probably one of the coolest and most underrated comments in this topic! Could you tell us more about this process?
- motohagiography 6y ago
  Thanks, it's way less cool when I write spectrogram instead of spectrograph. My mistake on that part. If you use those tools, it's really straight forward process of using the tools. I didn't know about Audacity until I had spent most of my time figuring out ffmpeg/ffplay. It was an amateur effort, but I've done a lot of security analysis.
kortex 6y ago
> In an era of fakes and deepfakes, we need tools like youtube-dl more than ever. I've been working on deepfake detection for the past two years. I use pytube3 and youtube-dl to scrape hours and hours of footage, not just youtube, but cspan, news sites, anything I can get my hands on. I have ~100 hours pulled so far, at multiple encoding rates. Facebook recently changed their API and now the best I can scrape easily is 240p. That's insufficient to detect the artifacts the models need to train on. I can only pull 1/3rd of cspan videos I can watch through the viewer because again, api changes. It's a constant cat and mouse. I'm not exaggerating when I say that if I don't have those wonderful folks out there keeping these tools and test suites up-to-date, this project is massively disadvantaged. I don't have any public release assets at the moment, but you can read about Siwei Lyu and Hany Farid, who are leading the charge in deepfake detection. http://www.cs.albany.edu/~lsw/ http://www.cs.albany.edu/~lsw/ https://futurism.com/expert-online-disinfo-covid-black-lives-matter https://futurism.com/expert-online-disinfo-covid-black-lives...

AnonsLadder 6y ago

That's freaking awesome

brokenmachine 6y ago

Did you post an article about it? I'd be interested to read it.

lostgame 6y ago

This is probably one of the coolest and most underrated comments in this topic! Could you tell us more about this process?

motohagiography 6y ago
Thanks, it's way less cool when I write spectrogram instead of spectrograph. My mistake on that part. If you use those tools, it's really straight forward process of using the tools. I didn't know about Audacity until I had spent most of my time figuring out ffmpeg/ffplay. It was an amateur effort, but I've done a lot of security analysis.

kortex 6y ago

> In an era of fakes and deepfakes, we need tools like youtube-dl more than ever. I've been working on deepfake detection for the past two years. I use pytube3 and youtube-dl to scrape hours and hours of footage, not just youtube, but cspan, news sites, anything I can get my hands on. I have ~100 hours pulled so far, at multiple encoding rates. Facebook recently changed their API and now the best I can scrape easily is 240p. That's insufficient to detect the artifacts the models need to train on. I can only pull 1/3rd of cspan videos I can watch through the viewer because again, api changes. It's a constant cat and mouse. I'm not exaggerating when I say that if I don't have those wonderful folks out there keeping these tools and test suites up-to-date, this project is massively disadvantaged. I don't have any public release assets at the moment, but you can read about Siwei Lyu and Hany Farid, who are leading the charge in deepfake detection. http://www.cs.albany.edu/~lsw/ http://www.cs.albany.edu/~lsw/ https://futurism.com/expert-online-disinfo-covid-black-lives-matter https://futurism.com/expert-online-disinfo-covid-black-lives...