Skip to content

Instantly share code, notes, and snippets.

Last active Nov 28, 2022
What would you like to do?
noise : relaxing ambient Brown noise generator (cf. white noise) | Linux bash script using sox | CogSci notes
#!/usr/bin/env bash
# bash 4.1.5(1) Linux Ubuntu 10.04 Date : 2019-01-02
# _______________| noise : ambient Brown noise generator (cf. white noise).
# Usage: noise [minutes=59] [band-pass freq center=1786] [wave]
# ^minutes can be any positive integer.
# Command "noise 1" will display peak-level meter.
# Dependencies: play (from sox package)
# Brownian noise, also known as Brown noise or red noise, is the kind of signal
# noise produced by Brownian motion, hence its alternative name of random walk
# noise. The term "Brown noise" does not originate from the color, but from
# Robert Brown who discovered Brownian motion. Brown noise is produced by
# integrating white noise. The sound is a low roar resembling a waterfall or
# heavy rainfall. We shall filter it through a band-pass, then add effects
# to mellow the output for ambience.
# Volume oscillation (amplitude modulation) is used to simulate artificially
# long ocean waves. Try some of the free online services listed at the end
# of this file to hear other types of oscillation.
# Our goal is to block out distractions for calm concentration.
# CHANGE LOG Get LATEST version from
# 2019-01-02 Fix sox warning by adding "-t alsa" appropriately for
# "play WARN alsa: can't encode 0-bit Unknown or not applicable"
# 2011-09-14 Add volume amplification to compensate for effects previously
# applied. Use peak-level meter to avoid clipping.
# 2011-09-13 Add bass and treble tone controls.
# at loud volumes, low frequency thumping may be annoying,
# so reduce gain on bass. Treble can be used to reduce
# harshness from the high frequencies.
# Adjust default wave for tremolo to perfectly cycle
# within the one-minute sample.
# 2011-09-12 Add end notes on the Cognitive Science aspects.
# 2011-09-11 Repeat use of one-minute segment to cut CPU usage by 95%.
# Fix tremolo to give very slow wave oscillation in volume.
# (Thanks to xguse for his gist at github.)
# Constant volume introduces tension psychologically.
# Code posted at
# 2011-09-10 First version based on 2009 article by Tom Swiss, and
# subsequent comments. See below for relevant portions.
# _____ Prelims
set -u
# ^ unbound (i.e. unassigned) variables shall be errors.
# Example of default assignment: arg1=${1:-'foo'}
set -e
# ^ error checking :: Highly Recommended (caveat: you can't check $? later).
# _______________ :: BEGIN Script ::::::::::::::::::::::::::::::::::::::::
repeats=$(( minutes - 1 ))
# ^increase for more volume oscillation, but suggest no higher than 0.20
# (and no lower than 0.0166667). Its value should consider the 60
# seconds duration of the repeated sample.
# ^your choice: 'white', 'pink', 'brown', 'tpdf'
# where tpdf stands for Triangular Probability Density Function (cf. dither).
# N.B. - white and pink noise have higher frequencies than Brown.
# ^CONSTANT one minute. (Format for specifying time length is hh:mm:ss.frac)
# ___ATTN___ We first pre-compute one minute of audio output to file,
# then playback repeatedly as necessary to satisfy minutes argument.
# This dramatically cuts CPU usage by 95% after the first minute.
# For DEBUGGING: "noise 1" shows the peak-level meter; also instant production.
if [ $minutes -eq 1 ] ; then
echo " :: Please stand-by... sox will 'play' $noise noise for $minutes minute(s)."
# FYI Channels: 2 @ 32-bit, Samplerate: 48000Hz.
play $progress -c 2 --null -t alsa synth $len ${noise}noise \
band -n $center 499 \
tremolo $wave 43 reverb 19 \
bass -11 treble -1 \
vol 14dB \
repeat $repeats
# # Previously published one-line versions misused tremolo:
# play -c 2 --null synth $len brownnoise band -n 1800 1400 tremolo 500 .1 reverb 50
# play -c 2 --null synth $len brownnoise band -n 2500 4000 tremolo 20 .1 reverb 50
# play --null synth $len brownnoise band -n 1200 200 tremolo 20 .1 reverb 20
# play --null synth $len brownnoise band -n 1200 200 tremolo 20 .1
# _____ ARGUMENTS explained via "man sox"
# "-t alsa" Needed to fix this sox warning (on 2019-01-02):
# "play WARN alsa: can't encode 0-bit Unknown or not applicable"
# Note: arg placement is critical, e.g. right after "play" will fail.
# -q, --no-show-progress
# Run in quiet mode when SoX wouldn't otherwise do so; this is
# the opposite of the -S option.
# -S, --show-progress
# Display input file format/header information, and processing
# progress as input file(s) percentage complete, elapsed time,
# and remaining time (if known; shown in brackets), and the
# number of samples written to the output file. Also shown is a
# peak-level meter, and an indication if clipping has occurred.
# -c 2
# Two channels. Without this, the output is not stereo.
# -n, --null
# This can be used in place of an input or output filename to
# specify that a `null file' is to be used. Note that here,
# `null file' refers to a SoX-specific mechanism and is not
# related to any operating-system mechanism with a similar name.
# Using a null file to input audio is equivalent to using a
# normal audio file that contains an infinite amount of silence,
# and as such is not generally useful unless used with an effect
# that specifies a finite time length (such as trim or synth).
# synth
# Some noise options: whitenoise, tpdfnoise, pinknoise, brownnoise.
# band [-n] center[k] [width[h|k|o|q]]
# Apply a band-pass filter. The frequency response drops
# logarithmically around the center frequency. The width
# parameter gives the slope of the drop. The frequencies at
# center + width and center - width will be half of their
# original amplitudes. band defaults to a mode oriented to
# pitched audio, i.e. voice, singing, or instrumental music.
# The -n (for noise) option uses the alternate mode for un-
# pitched audio (e.g. percussion). Warning: -n introduces a
# power-gain of about 11dB in the filter, so beware of output
# clipping. band introduces noise in the shape of the filter,
# i.e. peaking at the center frequency and settling around it
# Consider this for centering the band-pass...
# Freq (Hz) Octave Description
# 16 to 32 1st Human threshold, the lowest pedal
# notes of a pipe organ.
# 32 to 512 2nd to 5th Rhythm frequencies, where the lower
# and upper bass notes lie.
# 512 to 2048 6th to 7th Defines human speech intelligibility,
# horn-like or tinny sound quality.
# 2048 to 8192 8th to 9th Gives presence to speech, where labial
# and fricative sounds lie.
# 8192 to 16384 10th Brilliance, the sounds of bells and the
# ringing of cymbals. In speech, sound
# of letter "S" (8000-11000 Hz)
# Avoid the really low frequencies which will produce disturbing rumble.
# tremolo speed [depth]
# Apply a tremolo (low frequency amplitude modulation) effect to
# the audio. The tremolo frequency in Hz is given by speed, and
# the depth as a percentage by depth (default 40). Increasing
# the depth gives wider range between soft and loud volumes.
# reverb [-w|--wet-only] [reverberance (50%) [HF-damping (50%)
# [room-scale (100%) [stereo-depth (100%)
# [pre-delay (0ms) [wet-gain (0dB)]]]]]]
# bass|treble gain
# Boost or cut the bass (lower) or treble (upper) frequencies of
# the audio using a two-pole shelving filter with a response
# similar to that of a standard hi-fi's tone-controls. This is
# also known as shelving equalisation (EQ).
# gain gives the gain at 0 Hz (for bass), or whichever is the
# lower of ∼22 kHz and the Nyquist frequency (for treble). Its
# useful range is about -20 (for a large cut) to +20 (for a
# large boost). Beware of clipping when using a positive gain.
# When played loud, you may hear thumping bass lines in the
# case of brownnoise with effects. Reduce annoyance accordingly.
# vol gain
# Apply an amplification or an attenuation to the audio signal.
# Unlike the -v option (which is used for balancing multiple
# input files as they enter the SoX effects processing chain),
# vol is an effect like any other so can be applied anywhere,
# and several times if necessary, during the processing chain.
# The amount to change the volume is given by gain which is
# interpreted, according to the given type, as follows:
# if dB, then a power change in dB. When type is dB, a gain
# of 0 leaves the volume unchanged, less than 0 decreases it,
# and greater than 0 increases it.
# Beware of clipping when the increasing the volume.
# repeat count
# Repeat the entire audio count times. Requires temporary file
# space to store the audio to be repeated. [But where exactly?]
# _______________ "white noise" generator with sox [edited for code content]
# by Tom Swiss,
# January 2007, updated circa September 2009,
# included comments through September 2011
# Sox is "the Swiss army knife of sound processing programs." It includes sound
# generation capabilties for pure tones and white noise. "Pink noise" is
# also in sox's bag of tricks. After a bit of experimentation, I found the
# following shell script produced agreeable results:
# len='7:00:00'
# play -t sl - synth $len pinknoise band -n 1200 200 tremolo 20 .1 < /dev/zero
# __________ Comments
# Drew Haven: This beats the heck out of "cat /dev/urandom > /dev/dsp". The band
# filter is nice to take out the pops.
# gi1242: With recent versions of sox, things are a little simpler:
# play -n synth 60:00 brownnoise
# produces brown noise for an hour. (Replace brown with pink/white if you
# prefer. My baby sleeps best with brown).
# Tom Swiss: "Brown" in "brown noise" means Brownian motion. It's also called
# red noise. I learned something today, hooray!
# Adrien Beau, 30 January 2011: You can replace the "-t sl -" and "< /dev/zero"
# parts with the "-n" option, so your sox invocation becomes:
# ^= --null (for null file)
# play -n synth $len pinknoise band -n 1200 200 tremolo 20 .1
# The brown noise sounds the best in my opinion.
# Dennis Murczak, 5 May 2011: I adapted the line to a "my neighbor is having a
# party and I need to study" situation:
# play -c 2 -n synth pinknoise band -n 2500 4000 reverb 20
# The band pass is centered on human voice frequencies and wide enough to also
# cover most of the musical frequency range, without producing annoying
# high-pitched noise. The slight reverb adds a background/ambient quality for
# less distraction.
exit 0
# _______________ EOS :: END of Script ::::::::::::::::::::::::::::::::::::::::
# _____ Free ONLINE alternatives
# Simply Noise for white, pink and brown/red noise generator; uses Flash:
# (App is $0.99)
# [Flash consumes about 30 times more than our script in CPU usage!]
# PlayNoise for white, pink, and brown noise generator; uses Javascript/HTML5:
# has 33-second sample audio files containing perfect white noise.
# The randomness comes from atmospheric noise, which is more natural
# than the pseudo-random number algorithms. Such files could serve as
# input to sox for further signal processing.
# _______________ HOW WHITE NOISE WORKS by Saabira Chaudhuri
# WSJ 31 Aug 2011
# What people think of as "white noise" may actually be pink noise or brown
# noise or any number of other colors. Sound is associated with a color based
# on where it falls on an audio spectrum of high to low frequencies. White noise
# contains random sounds across all frequencies and "sounds very much like a
# hiss because everything is changing in every sample," says Daniel Ellis,
# associate professor of electrical engineering at Columbia University in New
# York. Pink noise, on the other hand, blends some high and lower frequencies,
# so it sounds like a hiss with a low rumble, he adds. Brown noise shifts to the
# lower end of the spectrum and sounds like rumbling.
# The most effective noise at blocking out other sounds is white noise because
# it covers the largest range on the spectrum, says Andrew Catellier, an
# electronics engineer at the Boulder, Colo.-based National Telecommunications &
# Information Administration, which publishes a glossary of sounds' color
# classifications. Distinguishing noises by their frequencies is a useful tool
# for scientists and engineers working on practical applications, such as
# building a cellphone system or an ultrasound machine. Sound is classified by
# its audible frequencies and associated with a color based on where it falls on
# the spectrum of high to low frequencies. White noise is unique in that it's
# random and includes all frequencies -- akin to how white light has all the
# colors in the spectrum.
# Calling sounds like rain or thunder white noise is somewhat of a misnomer, but
# the makers of downloadable apps and sleep machines use the term anyway. White
# noise and other soothing sounds, once mainly played on machines to aid
# nighttime sleep, are increasingly helping make daytime hours more serene.
# [White noise is a common synthetic noise source used for sound masking by
# a tinnitus masker.]
# After HeavyDutyApps released an app called Sleep Pillow Ambiance to help
# people sleep, it quickly realized that many customers used it during the day
# as well. "The usage varies from people who need help concentrating while
# working in noisy environments, commuters who need a break from train noise and
# travelers that need a peaceful environment," says Benny Shaviv, chief
# executive of the Westchester, N.Y.-based company. The $1.99 app has had more
# than 1.6 million downloads, says Mr. Shaviv. "By January we were among the Top
# 50 apps in the Healthcare and Fitness category in iTunes."
# Most popular are sounds from nature: rain, wind, waves crashing on the beach
# and crickets, Mr. Shaviv says. But the app also includes some unexpected
# sounds, such as cold drink with ice, brushing hair and horse running in field.
# Thunderstorm is the most popular downloaded noise.
# Developers of these apps say they frequently get requests for new sounds.
# Steven Jian, co-owner of Simply Noise, has received requests for the sound of
# passing cars and airport noises. Shaviv of HeavyDutyApps got a request for a
# sonar noise from a former sailor who served on a submarine. Todd Moore,
# founder and CEO of TMSoft, the maker of an app called White Noise, says he
# created a hair-dryer sound at one woman's request. "She told me that she could
# not sleep without listening to it and that she had burned [out] six hair
# dryers over the years."
# Daytime white-noise listeners say the sounds serve two main purposes: to block
# out distractions and lessen sounds that cause anxiety, such as sirens.
# "Certain types of noises can be relaxing," says Robert C. Fifer, director of
# audiology and speech language pathology at the University of Miami. White
# noise can be used to create a more relaxing working environment, masking
# sounds and promoting a sense of privacy, he says.
# One small study examined white noise in a classroom environment. The research,
# led by Goran Soderlund and Sverker Sikström of Stockholm University, looked at
# 51 students at a secondary school in Norway and found that those who normally
# had difficulty paying attention performed better when white noise was added to
# the classroom. The findings were published last year in the journal Behavioral
# and Brain Functions.
# The authors theorized that white noise boosted neural activity, helping the
# brain work more efficiently. The study predicted that white noise could help
# children with attention deficit hyperactivity disorder (ADHD) learn to focus
# on schoolwork better.
# "The effects of background white noise on memory performance
# in inattentive school children"
# Göran BW Söderlund1, Sverker Sikström, Jan M Loftesnes and EJ Sonuga-Barke
# Behavioral and Brain Functions 2010, 6:55 doi:10.1186/1744-9081-6-55
# Published: 29 September 2010
# Re: Brown noise, see
# vim: set fileencoding=utf-8 ff=unix tw=78 ai syn=sh :
Copy link

xguse commented Aug 18, 2013

I know that you do some help-like text in the comments but it would be nice if "noise -h" printed usage and option help text for random refreshing of one's understanding after a period of disuse. Otherwise the user (me) has to pull up the actual file and read through it.

Not a MAJOR deal but it is a convention that I think people expect.


Copy link

xguse commented Dec 10, 2014

I finally got around to messing with this bc i needed it and yet felt like procrastinating my own work... I have added named arguments and a callable help text.

Copy link

rsvp commented Jul 23, 2015

thanks @xguse, I will check out your changes.

Copy link

rgoodie commented Jan 23, 2016

Awesome and just what I'm looking for.

One trouble factor, when trying to run on a pi, I get the following.

pi@raspberrypi:~/code/brownnoise $ ./ 
 ::  Please stand-by... sox will 'play' brown noise for 59 minute(s).
play WARN alsa: can't encode 32-bit Signed Integer PCM
play WARN alsa: under-run
play WARN alsa: under-run

So I found this thread ( and it mentions adding "--single-threaded" to play. The under-run happens much less frequently.

No change needed to code. Just replying in hopes someone else is turning a pi into a night time sound machine.

Copy link

jezmck commented Mar 8, 2016

Thanks to @rgoodie, I am too.

Copy link

rsvp commented Mar 9, 2016

RELATED: Script for musically listening to binary bits:

So for example,
$ ./bin2music foo.jpg

On its own, will play random stuff from /dev/urandom:
$ ./bin2music

Have fun -- enjoy! [Please PR if you want to extend the scales.]

Copy link

antme0 commented Apr 22, 2017

Excellent script just what I was looking for. Many thanks!!!! Given me the idea to create a raspberry pi project for sleeping.

Copy link

rsvp commented Jan 2, 2019


  • 2019-01-02 Fixed sox warning by adding "-t alsa" appropriately for "play WARN alsa: can't encode 0-bit Unknown or not applicable"

Copy link

Thann commented Feb 1, 2021

I turned this script into a GUI:

Copy link

I actually use the -m length and noticed it is ignored, due to repeats=$( minutes - 1 ) being calculated with the default minutes before it is overwritten by the -m param.

Just moved that line down below param reading and it's fixed.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment