Company develops new technique to hide ID inside music files

I just posted the article Company develops new technique to hide ID inside music files.

A company called ‘SunnComm Technologies’ has licensed a new super-watermark that can be embedded inside music files. This new technique can also contain hidden data, video or software and can…

Read the full article here:  [](

Feel free to add your comments below. 

Please note that the reactions from the complete site will be synched below.

…and yet we see another technology that reduces quality! What’s the betting on a top notch hi-fi system (which I accept many of us won’t have) a code or signal like this will be audible? Let’s face it Crap-rovision screws up DVD on very large screen TVs and needs to be disabled. Just go suck your daddys willy guys and stop wasting everyones time and money with your shite…someone will only crack it anyway, i.e. find a way of removing it.

It’s actually just someone talking over the music with the information, LOL!

Heh I wonder if I could patent that! :slight_smile: Have some guy screaming all through the song “THIS SONG IS FROM MADONNA, DON’T STEAL IT!” It survives digital encoding, recording off the radio, and helps reduce piracy 'cause basically nobody wants it :slight_smile:

You just get a whole bunch of copies of the CD. Lets say, 12 of them. And then you rip all 12 of them. And then, you take each one, and randomly mix up the frames so you end up with one that has all differnt chunks of watermark from all the differnt copies. Tada, no more unique ID. Bye Bye watermark, you have been dispersed.

LoL Alox … And then you put on your own watermark that when read says screw you and displays a picture of middle finger with RIAA written on it =)

Alox, isn’t the digital watermark the same for all titles? I think mixing 12 different songs isn’t a great idea :9

Didn’t you read it? If the watermark is different for each copy of the disc, based on origin then alox’s would be a great idea, assuming you had an extremely accurate method of ripping that ripped bit-by-bit exactly the same data each time you ripped. Which would be difficult.

I had the great oppurtunity to meet the guy that developed this technology at University of Miami when I went for a tour of their Audio Engineering dept. earlier this year. (I am enrolling into their Audio Engineering program this coming year btw). I believe this technology was his Master’s thesis. This is amazing to hear of it actually being licensed and used! I was thoroughly impressed by his work even in the brief time I talked with him. I am going to have to recall exactly what I discussed with him so that I can post some more information on the subject as soon as I finish reading the article. This is very different from what is out there already for watermarking. It is embeded into the audio itself, does not affect the quality, and if I remember correctly, is not effected by lossy encoding. It can still be tracked. Great stuff if used correctly, and not selfishly. It can be used to embed text and other info into the audio stream, and if I recall, he was working on broadening the amount of data possible to be embeded into the audio when I was on the tour. Amazing!

Any ideas on how they can keep the watermark through an analog recording? Actually I have a bunch of question as to how this works and so on, but until I actually see it being used and not broken, I guess I can wait.

Nazrag: it doesnt need to be bit for bit at all. as long as you have several differnt sources (the more the better) and the watermark stays intact after ripping to WAV (its useless if it doesnt anyway) then when you mix them all up into a new, composite, WAV the watermark will be corrupted.

where exactly do you pick up analog recordings these days? very few things are released on cassette or lp, none with copy protection (i dont believe). im very curious as to how this extra 20 kilobits per second is woven into the cd.

Here is another idea for the hackers. If you can take 2-3 cd’s and rip them im sure since the watermark is embedded into the audio the diffrences can be detected and even if the encryption can not be broken it is possible to create a method to seperate most or some of the data from music and create a application to remove recognized data and/or make it all unreadable with random data recorded on. If audio is just waves then randomly making theses waves longer/shorter taller/smaller just slightly enough that it is not detectable by the human ear then watermark is currupted and unusable presuming that the watermark uses this same method to store data.

possibly the watermark is kept on an analogue recording by means of the waveform; all recordings have a wave form regardless of wether they are analogue or digital if they are copied identically the waveforms remain intact so it may be through this. :8

Nazraq: I only read this page, not the link.

Watermark!.. so what! You’re still listening to it… If the cops check your computer, you get busted anyway…watermark or not! Am I missing the point? Please let me know…I just don’t see the point of that watermark… if it stays there even after being converted to mp3 or ogg and one can not hear it… yeah… so what… :4 Sorry, I just DON’T get it! (I must be thick or something… :9)

Actually, they are trying to track down the Releaser with this method. Not the listeners of “illegal” rips of the original. Cut the evil from the root. They say about tracking down INSIDERS who rip and spread the albums BEFORE they’re oficially released. That’s the top of the problem.

Correct if I am wrong on two points. 1) The cd standard 44.1KHz was chosen because it supports the limit of AUDIBLE frequencies that the human ear can hear. So the watermark either has to be greater than 22.1KHz, or less than 20Hz. In the case that it’s greater than 22.1KHz, the signal will definately not be reproduced correctly, and will result in folded frequencies within the audible hearing range, creating distortion. In the case that it’s lower than 20Hz, it may very well create high currents through the amp and speakers, destroying your equipment faster. 2) MP3 and most lossy compression formats throw away “unheard” information, and only approximates the audible frequencies anyway. In this case, to be preserved the watermark must be within the audible frequencies Furthermore the amplitude has to be quite large, otherwise the approximaion will most likely discard the small ripple signal as noise. Congratulations guys for once again creating an unusable audio protection scheme. :slight_smile: I still can’t believe they pay you for that, and where can I sign up for your job. :g I have a job that is measured in both volume of output and timeliness. :frowning:

Whoaw war3peace… if they wanna do that for stopping insiders from releasing… that’s gonna be costy! If people don’t get their music before the official release… well they’ll get on the day of release. It’s just a matter of patience then… Also, whenever they find out who is the insider that released the material… it’s already spreading on the web and too late for them. So that guy gets fired until the next one gets caught! These guys in Miami should use their brilliant brain cells to better use. In my opinion, that’s just a waste of everybody’s time… … Maybe I just don’t give enough of a toss to be bothered. :8

Why does this company believe their protection is foolproof, they told us that with Charlie Pride’s Album and that made P2P networks. Its simple “Where There’s a Will, There’s A Way” Anxiously awaiting Tina Turner’s Album with copy protection from same company (The Early Sessions).