So you are saying that apps on your phone can access your microphone without your permission? Wouldn't you want to report that to Google or Apple or whoever made your phone's OS?
Also, how did your individual phone become relevant for the assumption that this is a widespread phenomenon?
Finally, it's great that you log your app activity, but you are aware that the scientist in the study I cited examined 17620 apps and found not a single instance of the app turning on audio and sending the data?
As I said, I'm sure companies will try. But what you're looking at there is a pitch deck presented to someone so they'd cough up money. That does in no way imply that this has been widely deployed without requiring user consent and therefore there are apps out there en masse, listening to smartphone users in a form of clandestine operation. It's basically the same thing as the patent for the old greentext Mountain Dew commercial meme at this point:
https://www.snopes.com/fact-check/sony-patent-mcdonalds/
In fact, the article clearly states that the data Cox Media Group uses comes from apps where users have agreed to grant the application permission to use their voice data, and that Cox Media Group was subsequently removed from the Google Ads program (a precaution for Google to save face).
This is actually not a thing. There is no evidence that smartphones constantly listen to people.
https://news.northeastern.edu/2018/07/06/is-your-smartphone-spying-on-you/
Also, this person is, well, for a lack of a better term, a porn actress. So this tweet might be a mix of bias, leading to the belief their phone is listening, but maybe also a bit of advertisement - which, don't get me wrong, is great. She did some stuff with Owen Gray, good stuff. Check them out, support their work.
That being said, it is certainly technically feasible. I'm sure there have been several, individual cases, and we will probably see new, "innovative" ways of companies spying on us.
It's actually not. There is no evidence that smartphones listening to users is a widespread phenomenon.
I have absolutely done that in the past, and I would do it again in a heartbeat. I can slice off exactly how much I want, getting a good feeling for the portion size. The noodles stick together, there is no unfurling and subsequent mess when transferring them to a dish. They'll loosen up a bit when heated, but stick together enough to offer a superior eating experience - just cut a piece off and consume.
That's a pretty normal picture if you have regular, social interactions. What's weird is posting that anywhere outside your circle of friends and the picture becoming the topic of public debate after one of the subjects allegedly shot someone.
That being said, and I say this as a married, retired old fuck - that guy definitely fucks.
They actually joke about it in the show, screaming "Aah, a crocodile", and it's just some woman working there. Same with the director being a lion, I believe.
The exhibit has turtles and penguins.
I'm a big fan of archive.org and I regularly look for manuals, but they don't show up in common searches as a source, so knowing that now is really helpful. Thanks!
I argued that exact same way with someone very close to me. Their answer was:
But those gods aren't real!
Nothing you can do if that last conclusion isn't there.
Some context:
After Elmo's venture into posting anime stuff on Twitter, this feels rather tame I guess. Or I'm just too jaded to care about this timeline any longer.
As someone who often lifts a finger, types out the first two sentences of a comment and then just resigns: thanks for the well-written comment.
I happen to work in machine learning. You are most likely referring to the Stanford Gyrophone paper. Given that the sampling frequency of the gyroscope sensor on typical smartphones is extremely limited, you can only get very low frequency content (Nyquist).
It wouldn't be possible for any human to process or understand the recorded signals, so the researchers trained a machine learning model on the recorded samples, with a very limited vocabulary consisting of only the digits from 0 to 9 and "oh".
If the model was not trained on the particular speaker (requiring annotated training data for that particular speaker, which would be almost impossible to get in the assumed scenario), the recognition rate was 26%. For a vocabulary of 11 words.
It's a nice proof of concept, and doubly so if tge CIA considers you a target, but otherwise it's not happening.