Is it possible to convert Ethernet packets into audio (in realtime or via capture file) and "listen" to the data that's...

Is it possible to convert Ethernet packets into audio (in realtime or via capture file) and "listen" to the data that's being transmitted and received?

Attached: 02_hyperelliptic_surfaces.flac.png (1920x971, 2.32M)

Change the file extension to .mp3 and play it back, if it fails, force no error.

>. Mp3

Why

Why would you even want this? It'll just sound like static.
If you still wanna go for it, my first guess would be to change the file extension from .pcap to .raw and tack on a header

I'm guessing its for art sake OR a neat way of auditorily listening for background data transfers while jacked into the interwebz

Attached: jacking_in.jpg (558x320, 22K)

>b-but m-muh flac!!!

Yeah but it'll just sound like noise.

I simultaneously am disgusted and supportive of your desire for a quasi-acoustically coupled dialup experience. But yeah you should be able to pipe the traffic as raw audio data into ffmpeg or some shit, then it'll package it up all nice as audio for ya

good idea, monitoring this thread.

The bursts you'll find in raw data be problematic. I imagine that if you take the full bandwidth, you'll end up with a ton of chirps so short they won't render to anything audible. Perhaps some product of packet frequency and length would produce something acoustically interesting? Think of stereotypical geiger counter.

>Think of stereotypical geiger counter.
nice pun

I envisioned just taking everything to know when it was active or not active. You could probably run it through grep, or use a utility that only grabbed higher layered packets TCP/UDP or even just HTTP or some shit rather than all the things at Ethernet protocol level (iirc there is some overhead there but it's been a long time since I looked at anything that low level). Presumably something like tcpdump would do the trick. Alternately you could take the entire bandwidth and like you said multiply by length of burst and try to get more meaningful or less noisy output. I wonder if you could do that though just by lowering the bitrate in the audio so that the super short blips would just slip between the cracks of the audio samples. Would probably be faster and let you do real time with less latency

If you really want to use the entirety of the incoming data, the only meaningful way to do it I can think of is buffering it and modulating the frequency based on the buffer size in such a way that you remain in real time (buffering time aside).

The mapping is arbitrary - You could surely translate it in a way that wasn't simply noise, hopefully something pleasant.
And then training yourself on known data, you could learn to pick out meaningful patterns/progressions - Listening to the stream you could have an idea of what was being transferred.
Like watching the matrix only actually doable - The human auditory system is really incredible at this sort of thing.
I'm not OP and have no idea if this is the intent, but I've certainly thought about this before - I think it's worth checking out.

I'm less worried about meaning rather than just an auditory alert that hey, my computer is actively using the network. Getting something meaningful I could iterate on just getting something at all, but that's secondary to me. Not OP though, dunno what his intent or purpose is

Air gap interception. It's theoretically been done, but not in a real attack afaik.

I never knew I wanted this

Yes, you can literally interpret anything as raw audio samples.
aplay /bin/* is always fun.

it would actually work better in this case to use a non-compressible audio format like WAV

Define better? Why would you not want it to be compressed?
t. audio brainlet

just got it working
get the ethernet card with tcpdump -D
then specify the interface with tcpdump -i 1 (where 2 is the card you want to use)
using tcpdump, ffmpeg and vlc (i dont have VLC in the path, so run this in the location where vlc is)

tcpdump -A -i 1 -q -n -C 30 -W 10 -U -s 0 | ffmpeg -f s16le -i pipe: -f mp3 - | ./vlc -

Attached: 1535503806251.gif (329x335, 382K)

If you want raw data to be converted as-is to audio, you wouldn't want compression to modify said data.

Ah, good point

>-f mp3
useless solution