How The NSA Converts Spoken Words Into Searchable Text

HardOCP News

[H] News
Joined
Dec 31, 1969
Messages
0
You'd think that something like this would churn out a million false positives a day.

Top-secret documents from the archive of former NSA contractor Edward Snowden show the National Security Agency can now automatically recognize the content within phone calls by creating rough transcripts and phonetic representations that can be easily searched and stored. The documents show NSA analysts celebrating the development of what they called “Google for Voice” nearly a decade ago.
 
It probably does, hence the reason the DHS is growing faster than the chinese economy and the national deficit combined.

More hits to filter and sort = job security at our expense.
 
I dunno why that kinda conversion would be expensive.

1. Dragon Naturally Speaking
2. Ctrl + F in a text file
3. Profit?
 
I cannot imagine how this type of system would function effectively without massive and far-reaching contextual descriptors.

....unless it's just a bulk data gathering system.
 
This is nothing new and reached the private sector years ago The new frontier is distinguishing stress patterns in voice and pattern trending through heuristics, machine learning, and big data analysis (if that's not enough catch phrases...).
 
This is nothing new and reached the private sector years ago The new frontier is distinguishing stress patterns in voice and pattern trending through heuristics, machine learning, and big data analysis (if that's not enough catch phrases...).

I have seen some interesting studies like this, but all of those were dealing with much better quality sound sources than cell phones :p
 
It probably does, hence the reason the DHS is growing faster than the chinese economy and the national deficit combined.

More hits to filter and sort = job security at our expense.

NSA has nothing at all to do with DHS, NSA is DoD.
 
I cannot imagine how this type of system would function effectively without massive and far-reaching contextual descriptors.

....unless it's just a bulk data gathering system.

It's easy, you use it to search for key text in intercepted communications on intelligence Targets, not against bulk data.
 
This is nothing new and reached the private sector years ago The new frontier is distinguishing stress patterns in voice and pattern trending through heuristics, machine learning, and big data analysis (if that's not enough catch phrases...).

Yup, exactly what the article said, 10 years ago.
 
NSA has nothing at all to do with DHS, NSA is DoD.

I'm not sure if you realize this, but any and all organizations that can have their names boiled down to three letters are all basically the exact same group of people doing the exact same things and whatever one of them does, it's just like the other ones doing it. So yeah, the FDA absolutely is the same as the ADA and the NES, GBA, CPU, and NSA.
 
ooohh. You're right.

I wish those TSA guys would get the hell out of my back yard.
 
ooohh. You're right.

I wish those TSA guys would get the hell out of my back yard.

The good news is that these guys don't count and are doing something different thanks to a lowercase letter i in their name.
CHiP%27s.jpg
 
I'm not sure if you realize this, but any and all organizations that can have their names boiled down to three letters are all basically the exact same group of people doing the exact same things and whatever one of them does, it's just like the other ones doing it. So yeah, the FDA absolutely is the same as the ADA and the NES, GBA, CPU, and NSA.


lol :p
 
Linda (with dry lips): Hey Al! Can you hear me? Al Kader! Where's the balm?

Albert Kader: I put the balm in the bag in the truck, outside the federal building.

Day ends badly for everyone involved :p
 
We use CallMiner to search collector calls to make sure they are in compliance (with federal and state laws). If not, they go bye-bye.
 
Back
Top