Trying to isolate sound in a certain frequency...

I’m currently playing an alternate reality game, and one of the clues we’ve been given is an audio file. We’ve already deciphered one code in the file, a simple morse code, but I’ve found what appears to be a second code within the puzzle.

The “code” consists of a beeping sound occurring in the 3950-4050Hz range of the audio file. I’ve managed to find it on an audio spectrograph, but I’d really like to isolate the sound for the ease of deciphering the code it contains. Is there a way to isolate sounds in just that range to make it easier to hear? I’ve googled the hell out of this, and I can’t find any info.

This is the forum where we are discussing the sound, including screenshots I made using Sonic Visualizer showing the beeps on an audio spectrogram.

The sound file itself can be found here, at the games trailhead:

You can (in Audacity 1.3) apply a high-pass filter lower than your work and then a low-pass filter higher than your work. Whatever is left is all the work in the middle with no competition from higher or lower frequencies.

Do you have any idea what you’re expecting to find? Is the frequency important? For example, if you say “440” to any musical person, they will know that’s the oboe tone at the beginning of the orchestra.

There are a number of critical frequencies like that. Ever read the magazine 2600? That’s taken from the telephone switch management frequency. I’m thinking if there are any well-known frequencies in your range.


The bottom one is a Flash presentation, so there’s no good way to capture it without a lot of work. Did you decode the Morse OK? I can almost still do that if I play it enough times. Something “R” something…


I sat through the whole thing on headphones. It may not be content. It be author. There was a sound design author that used to create these vast panoramas of mixed sound like that. Of course, I can’t remember his name.


Well, like I said I can faintly hear a series of pings, almost sonar-like. I’m trying to isolate them so I can work out what exactly they mean, and also so my fellow ARGers can hear what I’m hearing, as I’m having difficulty conveying what I’m hearing.

I can see them pretty clearly on Spectrograph, but they’re a little faint in places, so I was hoping by doing this I could better work out what they meant without all the background noise.

As far as the significance of the frequency, I couldn’t really tell you. My brother is more into the sound engineering side of things than I am, so I’ve gleaned what little knowledge I have of it from him, but unfortunately he’s partying it up at SXSW this week. I know the pings only occur at that frequency range. The game has also just started, so if that range has anything to do with the context of the ARG, I haven’t found out yet.

Feel free to come check us out at unfiction (if saying things like this isn’t allowed here, just lemme know). Like I said, the game is still in it’s first stage, so who knows what all of this means, and we could definitely use people who know what they’re talking about when it comes to sound. Usually the trailhead to an ARG gives a pretty solid clue as to what the rest of the game will be like, so if that’s true, we will probably need some expertise. And hey, it beats random surfing…

The beeps in the beginning translate to DCR. The poem that’s read directly after is called “A Thanksgiving Prayer” by William S Burroughs, from a spoken album called Dead City Radio, which we’re assuming is where the acronym is from. More than that, we don’t know, besides the estimated death toll will be six point five billion…there’s a facebook group, and you should be able to download the mp3 of the file here:

The “3950-4050Hz” peak appears to be the third harmonic of the 1350Hz peak …

BTW1 the metadata reads…

Track Title: DCR Prologue test

Comments: 00000000 00000210 0000076E 0000000000669E82 00000000 002531A3 00000000 00000000 00000000 00000000 00000000 00000000

BTW2 …

No I haven’t the foggiest what any of it means.

Here is a version of the track with a high pass filter gone through. You can more clearly hear the irregular dinging I’m describing throughout, but it’s still a little fuzzy in places. Can anyone think of a way to bring these dings out more so I can decipher their meaning, provided there is one? Or at least get them to show up clearer on an audio spectrogram?

The file can be found here:

NASA Quindar tones are close to 2600Hz …

(close to twice the 1350Hz peak)

Sputnik maybe …

I suspect they are generic “space” communication bleeps.

Sine waves (pure whistle) tend to only have one peak on the spectrum. Square waves or waves with clipping or overload damage have harmonic energy at odd numbered frequencies. 60, 180, 300, etc.

It’s one of the ways distortion analyzers work. Suck out the original pure sine wave and anything left is either distortion or noise (or both).

This page has a terrific movie illustrating how to get from sine to square waves by adding harmonics.


This filter plug-in may help.
Unzip the file (extract the .NY file from the ZIP file), then follow the instructions here to install (433 Bytes)

The result of 25 iterations of “Ultra-narrow” is monotonous, (that’s a compliment).

Thanks Trebor, I’ll take it as such :slight_smile: