Super User is a question and answer site for computer enthusiasts and power users. Join them; it only takes a minute:

Sign up
Here's how it works:
  1. Anybody can ask a question
  2. Anybody can answer
  3. The best answers are voted up and rise to the top

Hi everyone Im trying to follow a tutorial on generating truly random bits

How To Generate Truly Random Bits

This is the command from the tutorial but it does not work

rec -c 1 -d /dev/dsp -r 8000 -t wav -s w  - | ./noise-filter >bits

I know i can record my mic input using

rec -c 1 no.wav

this is the command i tried using

rec -c 1 -r 8000 -t wav -s noise.wav | ./noise-filter >bits

but i get

root@xxc:~/cc# rec -c 1 -r 8000 -t wav -s noise.wav  - | ./noise-filter >bits
rec WARN formats: can't set sample rate 8000; using 48000
rec FAIL sox: Input files must have the same sample-rate

I have complied noise-filter

noise-filter

I think the tutorial is using an older version of SOX and REC I'm using

sox: SoX v14.3.2 on Ubuntu 12.04 server

Can someone please help me ?

share|improve this question

This command line has too many outputs listed:

rec -c 1 -r 8000 -t wav -s noise.wav - | ./noise-filter >bits
                           ^file     ^standard output

For a command pipeline, the only output should be the - at the end. The rec command interprets the noise.wav parameter as an additional input, which will fail or produce bogus output. Try removing the extra filename (and other unnecessary/incompatible options):

rec -c 1 -t wav - | ./noise-filter > bits
share|improve this answer

per the tutorial you referenced (I'm following that one, too!) AND the latest-version SoX manpage (as of 9 Nov 2014), the complete, corrected command that worked for me was:

rec -c 1 -r 8000 -t raw -e signed-integer -2 - | ./noise-filter > bits

ent returned these values for a 49,152-byte file:

Entropy = 7.996356 bits per byte.

Optimum compression would reduce the size
of this 49152 byte file by 0 percent.

Chi square distribution for 49152 samples is 247.53, and randomly
would exceed this value 50.00 percent of the times.

Arithmetic mean value of data bytes is 127.5771 (127.5 = random).
Monte Carlo value for Pi is 3.147949219 (error 0.20 percent).
Serial correlation coefficient is -0.002336 (totally uncorrelated = 0.0).

This is pretty good quality random data!

share|improve this answer

Looks like your noise.wav is not at a sample rate that sox can record in, and sox requires both files to be at the same sample rate. You'll probably have to regenerate noise.wav at a higher sample rate.

share|improve this answer

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .