Berkeley CSUA MOTD:Entry 39766
Berkeley CSUA MOTD
 
WIKI | FAQ | Tech FAQ
http://csua.com/feed/
2025/04/15 [General] UID:1000 Activity:popular
4/15    

2005/9/20-21 [Uncategorized] UID:39766 Activity:nil
9/20    http://www.engadget.com/entry/1234000003058917
        Record keyboard sound, guess password with 90% accuracy.
Cache (5591 bytes)
www.engadget.com/entry/1234000003058917
Researchers at UC Berk eley claim they can just listen to keyboard taps and piece together a 96 % accurate reconstruction of English words typed and 90% of all random ly generated five-character passwords within 20 tries. The techniques us ed are relatively easy using a $10 PC microphone, open source spelling and grammar correction tools, and some custom code written by the resea rchers which will almost certainly end up on a torrent soon. This all wo rks cause like a congo drum, keys slapped at different points along the plate under the keyboard emit different tones. Apply a little statistic al learning theory and voila, your dirty little secrets are revealed. And here I just awarded the PSP Ceramic the Co olest Idea of the Week award, and now this? The fact that their technique appears to work, regardless of the keyboard being sniffed or the techique of the user, is what's truely remarkable here. shouldn't the software be released as a part of the project? though I would support torrenting it as a distribu tion method to save the school some cash on their bandwidth bill. fp I guess companies will have to start releasing extra quiet keyboards. But seriously, to my ear, on the powerbook notebook keypad, each key press sounds very much the same. a touch of macro can easily pro ve this method insufficient. how about lasers bounced off of windows to detect vibrations within the room, created by speech... i t's all been around for years and more likely decades. Posted Sep 15, 2005, 11:31 AM ET by jared You need 30 minutes on a P4 3GHz to process the ten minute recording befo re 'realtime' single keystroke grabbing can be activated. I wish the pap er had gone into more detail about what exactly a Mel-Cepstrum coefficie nt is. Y'all can build this at home using the paper as a guide, a copy of MARSYA S, an open-source audio processing toolkit that includes handy feature-e xtraction code, and a free Java machine learning toolkit like Weka. Posted Sep 15, 2005, 12:14 PM ET by Alex Man, too bad the researchers decided not to release the source code they used (not that it won't get leaked anyhow). But I was left a little conf used about how they got the initial sound-to-key recording. Did they hav e to figure out which sounds corresponded to which keys or did the progr am do it all for them? Plus, not all keyboards sound the same in the sam e places, slightly damaged keyboards might sound differently, although i f the code does everything it wouldn't make much difference. Posted Sep 15, 2005, 12:46 PM ET by Biochemlab When OLED Keyboards come out, it will be smart to make the letters appear in random places on the keyboard in order to enter a password... Perhaps that would be a good function button on an OLED Keyboard.... Posted Sep 15, 2005, 12:51 PM ET by matt and some day they'll probably invent some fandangled technology that can see with 100% accuracy what you type either save it on your computer in a hidden spot or send it covertly to the person who put it there. they'll probably call it a keylogger or some such t hing. the cool thing about a "keylogger" will be that if you can get close enou gh to put a microphone under somebody's keyboard you might as well just plant the program inside their computer and improve your accuracy by 10% . Posted Sep 15, 2005, 1:12 PM ET by Steve I think the real processing power will have to be in the training phase o f the algorithm. This technique really opens up the possibility of makin g a "virtual keyboard" on any surface, too. Have a small switch to "acti vate" the virtual keyboard on your coffee table, and then just type on t he coffee table! Posted Sep 15, 2005, 1:33 PM ET by slyecho thats why biometrics are becoming more and more necessary. But then again, they could always plop out your eyes and chop off your fi ngers if they needed access to secure systems reeealy bad. Posted Sep 15, 2005, 1:49 PM ET by Mike Abdullah What about the possible use in powerless wireless keyboards? If the technology could be improved enough, surely one could just have a keyboard consisting of nothing but the keys. Once the computer has "lear nt" your keyboard it can tell what you're typing. Posted Sep 15, 2005, 2:03 PM ET by Dave "and some day they'll probably invent some fandangled technology that can see with 100% accuracy what you type either save it on your computer in a hidden spot or send it covertly to the person who put it there. they'll probably call it a keylogger or some such thing. the cool thing about a "keylogger" will be that if you can get close enou gh to put a microphone under somebody's keyboard you might as well just plant the program inside their computer and improve your accuracy by 10% . Sure, you could plant a tape recorder or whatever, though your point m akes tons more sense if a desktop is involved. This would also make it easier to record keystrokes from a distance. You can use a laser mic t o record the keystrokes, then decode them at your leisure. org/wiki/Dvorak_Simplified_Keyboard) Add your comments Please keep your comments relevant to this blog entry: inappropriate or p urely promotional comments may be removed. To create a live link, simply type the URL (in cluding http://) or email address and we will make it a live link for yo u You can put up to 3 URLs in your comments. Line breaks and paragraphs are automatically converted no need to use <p> or <br> tags. Your name (required): Your email address (required, will not be shown to the public): Your sites URL (optional): Do you want us to remember your personal information for next time?