Is DARPA Calling for Improvements to ECHELON’s Voice Recognition Systems?
February 16th, 2010I’ll just plonk my entire commentary from U.S. Allegedly Listened In on Calls of Americans Abroad—I Know, You’re Shocked into this post:
This is probably the biggest not-news-to-you-or-me post ever to appear on Cryptogon. What is described below has been standard operating procedure for NSA for decades. What seems strange, though, is the amount of mundane traffic that is winding up in the queues of the human analysts. This could mean a few things:
1. The ECHELON dictionary systems aren’t very good and are generating a lot of false positives.
2. The net is being cast much wider on purpose, using more mainstream terms to task the dictionary systems.
3. More individuals are being targeted for routine surveillance, which may result in more human transcriptions of the raw intercepts.
4. All of the above???
Also, I find it interesting that the allegations only cover calls originating in the Middle East and the Green Zone in Iraq. Is the point for us to think that we’re not being monitored when we make international calls from other areas???
I suppose that some imbeciles might reach that conclusion from reading this.
And now…
Via: Register:
So it’s interesting to note that Pentagon boffins have now stated that perhaps the most intriguing reputed capability of Echelon – the ability to automatically pick out words of interest and flag that conversation up as important to its human masters – doesn’t work. Or anyway, it only works on good, clear lines: a noisy or degraded signal frustrates it.
The news comes as part of a solicitation from the Pentagon crazytech bureau, DARPA, in which the maverick military mayhem mavens request assistance with building a Robust Automatic Transcription of Speech (RATS) system. According to DARPA:
Existing transcription and translation and speech signal processing technologies are insufficient for working with noisy or degraded speech signals that are of importance to current and future Department of Defense (DoD) operations. Currently, there is no technological solution [our emphasis] which effectively addresses this kind of noisy and distorted speech signal, so operational units are forced to allocate significant human resources for this task.
One should note that America’s feared National Security Agency (NSA, generally thought to be in charge of Echelon) is actually an arm of the DoD, not a civilian organisation.
DARPA says that the proposed RATS system should be able to tackle noisy audio signals and tell on its own whether they are speech or something else such as music. It should then be able to identify the language being spoken, and tell whether the speaker is a person of interest using voiceprint technology. Finally, the RATS software should be able to “identify specific words or phrases from a list of items in the language being spoken” – just what Echelon is supposed to be able to do already, only DARPA assure us that no such tech exists. Or anyway, none able to tackle a noisy signal.
And how exactly do they translate/transcript swiss-german? Or eastern (gibberish-) german? or some mumbojumbo-marketing-speak that doesn’t make any sense at all?
Do they just hoard the data until needed against somebody? I doubt that a system who does that calculates with everything except a immanent collapse.
I liked the sentence: The news comes as part of a solicitation from the Pentagon crazytech bureau, DARPA, in which the maverick military mayhem mavens request assistance with building a Robust Automatic Transcription of Speech (RATS) system.
It is so an honest description.
For random words its fine. A string of simple words would trigger a flag. If you have any connection to anyone of interest or are in an interesting part of the world ect.
The other side is known numbers and known voice prints.
Once your in the system, its easy. Somebody sits in a small space and does the “translate/transcript” as the people are known or linked to a known person.
Re “Do they just hoard the data until needed against somebody?”
Not really, ‘everything’ from a set of optical and sat links enters in real time. Its then sorted for a list of words or known people.
A tiny quality fraction is encrypted and fed back to the USA from Aus, NZ, Canada, UK ect every day.
That was the classic idea, suck in all, sort, filter, send back, sort, watch, pass on for further action by other groups within the US gov.
As the NSA is now linked to Google legally and in the ‘fly over states” within the US internally and legally – “hoard the data until needed against somebody” is looking to be more interesting. Linking back to the US from distant downlinks is not an issue, they are in your backyard and can just sit on it all.
Smart people meet in person in pools, ocean, away from anything electronic and use their PC’s for games and sports.
Phones are used for family and friends or short cryptic once only bursts.
Everybody state and state trained actor knows not to trust anything networked now, so you have to wonder who they are listening in on and expanding for. Would an ex spy or merc selling to the ‘private’ sector not advise the same???