White noise can be produced by a stream of random numbers that is then listened to as an audio signal. "White" refers to the even distribution of wavelengths in white light, with a particular meaning in the audio or DSP sense -- that the power of the noise is distributed evenly over all frequencies, between 0 and some maximum frequency which is typically half the sampling rate. For example, white noise at a sampling rate of 44,100 Hz has as much power between 100 and 600 Hz as between 20,000 and 20,500 Hz. To human ears, this seems bright and harsh.
In the natural world, there are many physical processes that produce noise with what is known as a pink distribution of power. Pink noise has an even distribution of power if the frequency is mapped in a logarithmic scale. For example, there is as much noise power in the octave 200 to 400 Hz as there is in the octave 2,000 to 4,000 Hz. Consequently it seems, human ears indicate that this is a natural, even noise.
You can filter white noise to generate pink noise. A DSP or analogue electronic low pass filter with a -3dB/octave (-10dB/decade) response when fed with white noise produces pink noise, within certain limits of accuracy.
You can use many different filter designs for this process. This example provides a working utility for those who want to generate pink noise without having to create their own version. Sound output is provided, demonstrating the difference to human ears between pink and white noise.
Example code from the Example Code Exchange in the NI Community is licensed with the MIT license.
