White noise can be used to protect against audio surveillance. Counter-surveillance white noise generators are available on the free market, including desktop and portable models, and a transducer that ‘injects’ white noise into walls and ceilings. You can also build one yourself in c++ (1, 2), Java (1, 2, 3), etc.. Whether buying or building, you probably want to experimentally verify/falsify the actual effectiveness of the white noise against surveillance (i.e., can the observer still filter out what is being said) under various circumstances (voice dynamics, characteristics of observer’s microphone, dish, filtering capabilities) before using it in practice.

Now, on the [TSCM-L] mailinglist, Technical Surveillance Counter Measures (TSCM) expert James M. Atkinson shares this interesting alternative to white noise as measure to protect against audio surveillance:

Good Evening Folks,Here is a little sample of some of my TSCM Audio Countermeasures projects.

http://www.tscm.com/Babbler1.wav [link is now dead]

It is only 50 minutes long and is 1 GB is length, so it is pretty small.

Put it on your iPad or iPad and plug in some powered speakers and you will destroy the ability of anybody to eavesdrop on you, and you will bring even the most sophisticated eavesdropping system right down to their knees. Then set the track for an endless loop, and either have your confidential meeting or start running your sweep.

You will need to be very close to the person you will be speaking with, then you place this audio on speakers that fill the room with sound, and you can watch the eavesdroppers heads explode, and their eavesdropping hardware essentially burst into flames. Even the most sophisticated adaptive filtering hardware and software the FBI and CIA used can not defeat this audio jamming track when you use it right.

This is actually a fairly narrow babble tape, but it will get you started. To mix down a proper tape I also need a 15-30 audio recording of each person who will be at the meeting reading form a news paper or book non stop (two hours of this is ideal). Then I will take each person create four vaiations of each persons voice to compensate for normal pitch changes and mix this into a single track for them. Then I normalize the track in amplitude to match all the other tracks that I create for the other people on the tape in the same way. Each track is slowly build one at a time, and to include the voices of every person who will be in the meeting on the babble tape.  This is called “Active Speaker Tracks” then I do the same with people who will not be at a meeting called the “Inactive Speakers Tracks” and bring this in.

Then I bring in what is walled the “Background track” which is what you hear in this audio file (this is the base of background track). This base track also contains “pips” of coded time bursts that are used to calculate the distant to  a bugging device and this is hyper efefctive. I have also placed sub-sonic and hyper-sonic signals and short periodic peaks that will shut down the automatic gain control circuits in many eavesdropping of bugging device.

This is strictly a demo of my work, for your enjoyment.

When I create these for clients it requires several days of work as I want a 4 hour file, that loops, and the file will have 64 channels mixed down to single MP3 or .wav file, I also create babble tape for one specific customer at a time, and often they will purchase a half dozen different files for different  situations, and that are devastating effective.

I give away single base file like this, and if a customer wants one of their own then I change $2500 per 4 hours, on a 64 track custom babble tape, including the voices of the people they are trying to protect, so get in touch with me if you woudl like a customer version of his made for you.

This file is dangerous, very dangerous, use it with grade caution

Enjoy, and let me know if you need more.

Sounds like a fun idea for experimentation.

UPDATE 2012-06-18: babble-tape.pdf (mirror) was posted on Cryptome. It states: “NOTICE IS HEREBY GIVEN that on the 24 day of February in the year 2012 at 8:30 am, or as soon thereafter that the matter may be heard at Dept. 1B of the aforementioned court. The defendant, Kaushal Niroula, A Propia Persona will move and hereby moves this court for an order of this court compelling Stanley Sniff Sheriff of the County of Riverside and his subordinates and all incarceration facilities operated within his jurisdiction through his designates to allow the use of the ‘Babble file’ on his computer and a USB speaker during his visits either with Court appointed assets or Investigators or attorneys while incarcerated in the Riverside County Jail in Indio or anywhere else.”

UPDATE 2012-01-24: James M. Atkinson states the following:  “In a recent laboratory test, this babble mix utter blew away all other forms of white noise generators and white noise masking systems, and Acoustic Noise Generators, just blew them away, and this is actually a pretty simple one, it I make one that is specific to a person, place, or such it can be way more confounding to the spy. For a fully tweaked out babble file I include the persons voices who will be in the meeting that you are trying to protect, and this is lethal to the eavesdropper, because I do all kind of cute things to the audio to foul up pitch, speed, and tempo, and I  when I mix it down to two tracks the tweaks can not be heard, but you sure can see then on test equipment.
When properly tweaked one of these files will become an “Anti Adaptive Audio Filtering” file, not this one specifically, but one which includes the actual speakers voice, and will destroy and derail even the most determines eavesdropper. Of you just need a general purpose babble file the one listed in the link is perfect, if you want one tweaked out properly to you voice or to the meeting, or even to the tone and dimensions of a specific room I can do this as well. This is a dangerous file, all you need to make it work is a cheap MP3 Player, and some powered speakers and for under $75 in electronics you get performance that exceeds what a $4000 Acoustic Noise Generator is capable of. The U.S. Government taught me well.”