Comment by BeetleB
10 hours ago
> So they are claiming that ACR is done entirely through audio processing, and no visual data from the screen itself is used? That...doesn't seem plausible.
It apparently is.
Do you know the Nielsen Ratings? That's exactly how they know what people are watching these days. I've been contacted by them to be one of their subjects. They ship you a device that listens and recognizes based on content. They look for certain "fingerprints" in the audio and match against that - it's not capable of recording our conversations, etc (or so they claimed).
Is that how they do it lately?
I participated once or twice a good number of years ago, and it was kind of a pain: It involved keeping a paper ledger of what was watched, and when. Correspondence was all handled by regular mail.
(As I recall, they included a small amount of cash ~every time they sent more correspondence and this was a primary motivator to keep going with it.)
The Shazam app on my wife’s iPhone can identify TV shows and movies very quickly, sometimes nearly instantly. Not sure how it does that.
It is pretty neat and several resources online explaining it but it chunks the recoding and does a hash of the fft of each frame i think which reduces the data a lot. The server has a pre hashed database https://hackernoon.com/how-shazam-works-in-a-nutshell
I forgot the part about finding the dominant frequency in each chunk
I know a guy who made a Shazam-like phone app that can listen to a few seconds of noisy audio and make a fingerprint that can be used to quickly ID those specific few seconds out of a pre-fingerprinted archive of an enormous amount of audio (zillions of hours of TV). Making the fingerprint requires a tiny amount of very smart code (in C with no dependencies). But, the fingerprint is not useful for understanding audio that's not in the archive.
TV shows are completed and in the can well before air date. It is entirely plausible that the interested party in having the content identifiable by Shazam like services submit whatever needs to be submitted for new content
I have also wondered about the possibility of media emitting some kind of hidden identifier. Can normal speakers emit noise outside of human hearing that could contain a unique id? Would not even need to be a lot of data that could be broadcast.
4 replies →
Signals analysis is pretty refined and not THAT compute intense.
This seems like a credulous argument based on no facts you have provided.
If you explained how the box was hooked up, it would be easier to judge whether there was a chance you were correct.
As it is, your description isn’t even clear whether or not the box contains a microphone.
Nielsen is global Company that is used for pricing ads on TV stations around the world. Their way of doing it from what I know personally is that they sign up people to put this ACR boxes voluntarily and then based on extrapolation of data - they can tell you pretty close what was the viewership across demographics. Similar to how election polls are done - you just need a good sample.
It's been done for at least good 15 years.
nielsen's ACR method is embedded into millions of devices and vendors. No need for a microphone, it's processing the audio signal being sent out. A microphone would be a wasted conversion..
(your car infotainment also likely has nielsen tech in it most likely, as they bought arbitron and gracenote)
Notably, the TV was caught making audio recordings and transcripts of conversations, and sending them to LG, independently of all this.
LMGTFY was made for people like you.
You know you can look it up if you are skeptical, right?
Also, they didn't claim that.
It's how things were always done. The content from broadcasters has/had audio watermarks...frequencies which humans can't hear.
As far as I know, watermarks never took off because the technical complexity and lack of standardization. Nielsen researched and developed ACR technology but it never required watermarks.
When Nielsen first started collecting data (1950s), they handed out diaries to a group of around 5 000 selected households. Nielsen would then collect these diaries and extrapolate viewing patterns across the entire population.
After that came phone surveys. Next step (1980s) was the "people meter", a small box attached to the TV that could automatically detect which channel was tuned and which household members were in the room.
In the 1990s the set-top box made all of that redundant as the cable and satellite providers could directly see which channel every single box was tuned to at all times.
I have a friend who got a Nielson device recently and they were told it relies on watermarks. Their website says a combination is used: https://www.nielsen.com/insights/2023/how-to-measure-tv-audi...
> watermarks never took off because the technical complexity and lack of standardization
Both audio (sports) and video (netflix et al) watermarks are alive and well. They are used to trace piracy (ie account ID), not for content ID (at least AFAIK).
Broadcasters: the cutting edge of streaming information is ours! Now let's make it so everyone unknowingly broadcasts as much info back to us as possible!
Poor Farnsworth. He just wanted edutainment, man.
A long time ago (~1998) I did some consulting for VEIL and Neilson was definitely talking to them...
https://en.wikipedia.org/wiki/Video_Encoded_Invisible_Light
Regardless of it being how it always was done, such a finger print can easily be injected into LG's advertising ecosystem to infer everything about your demographic and what your interests and income might be and hit you with more specialized ads.
What we watch is not OK to be fed ads, especially if I wasn't using their service to being with and its monitoring the audio from any input to do this.
In this way, smart tvs and smart home devices are effectively hostile devices behind your router at home.
Yes. It’s also horrible that they have microphones in your home listening to your conversations. LG knowing what I’m watching is awful, but it seems even worse for LG to listen in to the conversation I have with my loved ones about it after. I don’t want to be under surveillance in my own home.
The good news is there is no good reason to plug the TV into the internet.
3 replies →
Both are unacceptably egregious and done for zero benefit of the customer and a disproportionate and asymmetric benefit for the manufacturer.
I've been trying to learn how to secure LG TVs, and saw that some people do seem to jailbreak them withan open version of webOS.
1 reply →
[dead]