> ## Content Index
> Fetch the complete content index at: https://soundforgepro.com/llms.txt
> Use this file to discover other available public pages before exploring further.

# White Noise Generator: Play White or Brown Noise Online
- URL: https://soundforgepro.com/white-noise-generator/
- Published: 2026-10-28T15:06:19.000Z
- Updated: 2026-10-28T21:27:22.000Z
- Description: White or brown noise as a 60-second stereo loop made in your browser, with a softness filter, a sleep timer that fades out and a WAV save. Below the player: how the colors differ, which one covers voices or rumble, how loud machines get in published tests, and safe distance for a baby.
- Author: Erick Finn
- Tags: Audio Tools, Audio Fundamentals

White noise has equal power in every hertz, so it sounds like hiss or radio static. [Brown noise](https://soundforgepro.com/brown-noise-generator/) loses 6 dB per octave as pitch rises, which puts most of its energy low and makes it sound like a rumble or a distant waterfall. The player below makes either one on your device as a gapless 60-second stereo loop, with a softness filter, a sleep timer and a WAV save. Keep it only as loud as the sound you want to cover.

Browser tool · white and brown noise

## Play white or brown noise

Signals are made on this device. No microphone, no upload. Each color is a 60-second stereo loop with separate left and right noise.

**Lower system volume before Play.** Tool level starts at 5%, caps at 25% and is not dB SPL. For sleep, keep the speaker across the room, never in a crib, and only as loud as the sound you want to cover.

**Flat per Hz**White noise **\-6 dB per octave**Brown noise 

Spectrum shape 

**White**0 dB/oct

1001k10kHzdB 

Even hiss, flat per HzSoftness offTimer off

Playback**Stopped**

Press Play. Color, softness and timer can change while it plays.

PlayStop

Softness · low-pass

Off8 kHz4 kHz2 kHz1 kHz

Sleep timer · 60 s fade-out

Off15 min30 min45 min60 min90 min

**Keep it playing:** leave this tab open and the device charging. Desktop Chrome keeps a sounding tab running; on iPhone and iPad, locking the screen or switching apps pauses web audio, so keep the screen on or save the loop below.

Tool level

5%

Loop length

30 s60 s

Sample rate

44.1 kHz48 kHz

Save WAV loop 

16-bit stereo WAV, current color and softness: 11,520,044 bytes (11.5 MB). Saved at a fixed level with peaks below -1 dBFS, so start your player quietly.

Ready. Pick a color, then press Play.

White: Gaussian noise, flat per Hz. Brown: white noise through a leaky integrator, falling about 6 dB per octave above about 60 Hz. Colors are matched by K-weighted loudness. Not calibrated output or an SPL reading. Press Escape to stop.

## How to use the white noise generator

1. **Lower the volume.** Turn the system volume down before the first press. The tool level starts at 5% and stops at 25%, and it's a digital level, not a reading in dB SPL.
2. **Pick a color.** White for even hiss that covers voices, Brown for a low rumble. Switching while it plays crossfades over half a second instead of cutting.
3. **Press Play.** The noise fades in over 0.4 s. Stop, or the Escape key, fades it out in a quarter of a second.
4. **Set softness.** Off plays the full band. The 8, 4, 2 and 1 kHz settings roll off everything above that point at 12 dB per octave, which takes the edge off white noise and leaves the bass alone.
5. **Set the timer.** Pick 15, 30, 45, 60 or 90 minutes. The countdown shows under the readout, and the last 60 seconds fade to silence before playback ends.
6. **Save a WAV loop.** Optional: choose 30 or 60 seconds and 44.1 or 48 kHz, then press Save WAV loop. You get a 16-bit stereo file of the current color and softness, built so its end runs into its start without a click; set your player to repeat one track.

Color, softness, level and timer all change while the noise plays. The level slider moves smoothly, and a new timer setting starts counting from the moment you pick it. Your color and softness are remembered on this device; the timer and level start fresh each visit.

## White vs brown vs pink noise

The three colors differ in one thing: how power changes as pitch rises. The [University of Illinois acoustics lecture notes on noise](https://courses.physics.illinois.edu/phys406/sp2017/Lecture%5FNotes/P406POM%5FLecture%5FNotes/P406POM%5FLect15.pdf?ref=soundforgepro.com) define white as flat per hertz, pink as falling with 1/f and brown with 1/f², a slope of 20 dB per decade. Per octave that's 3.01 dB down for pink and 6.02 dB down for brown.

__Slopes from the definitions; the 250 Hz shares are measured on this page's own generator at 48 kHz.__
| Property                       | White                                         | Pink                                                          | Brown                                    |
| ------------------------------ | --------------------------------------------- | ------------------------------------------------------------- | ---------------------------------------- |
| Power per hertz as pitch rises | Flat                                          | −3.01 dB per octave                                           | −6.02 dB per octave                      |
| Power in each octave band      | +3.01 dB per octave                           | Equal in every octave                                         | −3.01 dB per octave                      |
| Often described as             | Static                                        | Rain                                                          | A waterfall                              |
| Share of power below 250 Hz    | 1.03%                                         | Not made on this page                                         | About 93%                                |
| Best fit                       | Covering voices and other mid and high sounds | Speaker and room checks; best sleep-study record of the three | Covering low, steady rumble without hiss |

The descriptions come from [Nigg and colleagues' 2024 meta-analysis on noise and attention](https://pubmed.ncbi.nlm.nih.gov/38428577/?ref=soundforgepro.com), which says white noise sounds like static, while pink and brown may sound like rain and a waterfall. The 250 Hz row explains most of the difference you hear. White noise spreads its power evenly per hertz, so nearly all of it lands in the upper octaves, which are much wider in hertz. Brown puts about 93% of its power below 250 Hz.

Pink sits between them, with equal power in every octave. It has its own [pink noise generator](https://soundforgepro.com/pink-noise-generator/) on this site, and it's the better choice for checking speakers or a room by ear, since each octave of the speaker gets the same share of the signal.

## Which noise color to use for sleep, focus and masking

Choose by the sound you want to hide. Masking works where the noise has energy, so a low hum needs low energy and a conversation needs energy in the speech range. The site's [audio frequency chart](https://soundforgepro.com/audio-frequency-chart/) places speech detail between 500 Hz and 2 kHz and intelligibility between 2 and 4 kHz. White noise has plenty of power there. Brown keeps only about 7% of its power above 250 Hz.

- **Voices through a wall or an open office.** I'd pick white, with softness Off or at 8 kHz, because brown leaves the speech range nearly empty.
- **Traffic, a fridge, a neighbor's bass.** Brown fits low, steady sounds. Its energy sits where they do.
- **Falling asleep in a mixed-noise room.** I'd start with brown and switch to white only if voices still come through. Brown has almost no hiss, and both colors play at matched loudness here, so the switch doesn't jump in level.
- **Focus.** Pick whichever color hides the background at the lowest level. The evidence for any color is thin, as the research section below shows.

The playback device changes the answer for brown. The [bass test](https://soundforgepro.com/bass-test/) notes that phones, laptop speakers and compact Bluetooth speakers usually run out of bass several steps earlier than full-size speakers. With 93% of brown's power below 250 Hz, a phone speaker plays mostly its thin upper edge, and turning it up to make up for that pushes the small driver toward distortion. My pick for most bedrooms is brown on a speaker with a real woofer, set across the room, or white on a phone.

## How loud white noise machines get: published measurements

Published measurements agree on one point: most machines can play far louder than a room needs, and distance does most of the work in bringing the level down. Twenty machines measured by [No Sleepless Nights on each one's loudest sound, averaged over three readings at 1 m](https://www.nosleeplessnights.com/best-white-noise-machine-reviews/?ref=soundforgepro.com), put the median at about 71 dBA, with a range of 54.7 to 79.9 dBA; half of them reached 70 dBA or more in a room that measured 31.3 dBA with nothing playing.

__Published sound-machine output measurements: one 20-machine product test, two studies of 14 devices each, and one literature review.__
| Source                               | Devices                      | Conditions                                                                    | Result                                                                                 |
| ------------------------------------ | ---------------------------- | ----------------------------------------------------------------------------- | -------------------------------------------------------------------------------------- |
| No Sleepless Nights review, 2025     | 20 sound machines            | Loudest sound, mean of 3 readings at 1 m, A-weighted                          | Median 71.0 dBA (mean 70.8), range 54.7–79.9; 10 of 20 at 70 dBA or more               |
| Hugh et al., Pediatrics, 2014        | 14 infant sleep machines     | Maximum volume at 30, 100 and 200 cm, corrected for a 6-month-old's ear canal | 14 of 14 above 50 dBA at 30 cm; 3 of 14 above 85 dBA                                   |
| Hong et al., 2021                    | 8 machines and 6 iPhone apps | Maximum and minimum volume at 10, 30 and 100 cm                               | 9 of 14 above 85 dBA at 10 cm at maximum; none at 30 or 100 cm; none at minimum volume |
| De Jong et al., Sleep Medicine, 2024 | Review of published data     | Maximum volume                                                                | White noise machines can exceed 91 dB                                                  |

Put together, the numbers show where the risk is: maximum volume, close to the ear, for hours. The [2021 measurements by Hong and colleagues](https://pubmed.ncbi.nlm.nih.gov/33992973/?ref=soundforgepro.com) make the distance effect plain, with 9 of 14 devices over 85 dBA at 10 cm and none at 30 cm. Six of those devices were iPhone apps, so the same distance rule applies to a phone.

For a sense of scale, the [WHO's safe listening guidance](https://www.who.int/news-room/questions-and-answers/item/deafness-and-hearing-loss-safe-listening?ref=soundforgepro.com) gives adults up to 40 hours a week at 80 dB and only four hours at 90 dB, and it suggests keeping device volume at no more than 60% of maximum. Eight hours a night comes to 56 hours a week, more than the 40 hours allowed at 80 dB, so a machine playing near 80 dBA at your pillow all night is past the guidance.

My rule is simple. Set the level by ear against the sound you're covering, stop raising it once that sound fades into the noise, and put the speaker across the room from your head. The slider here can't tell you dB SPL. For a number, a sound level meter app at pillow distance gives a rough reading; Hong's team took their readings with NIOSH's SLM app.

## What the research says about white noise for sleep and focus

The sleep evidence is weak. A [2021 systematic review by Riedy and colleagues](https://pubmed.ncbi.nlm.nih.gov/33007706/?ref=soundforgepro.com) of 38 articles rated the evidence that continuous noise improves sleep as very low quality under GRADE, and it warned that continuous noise may also negatively affect sleep and hearing. A [2022 review by Capezuti and colleagues](https://pmc.ncbi.nlm.nih.gov/articles/PMC9163611/?ref=soundforgepro.com) of 34 studies with 1,103 people found no strong evidence that sound helps sleep. A third of its white-noise studies and about four in five of its pink-noise studies reported better sleep, and none of the 34 reported adverse effects with short-term use.

Focus has one solid meta-analysis. Nigg's team found a small benefit of white and pink noise on lab attention tasks for people with ADHD or ADHD symptoms (g = 0.249 across 13 studies and 335 people), and a small negative effect in comparison groups without ADHD (g = −0.212 across 11 studies). They found no studies of brown noise at all, so claims that brown noise sharpens focus have no trial behind them yet.

I'd use noise as a masking tool and nothing more. It hides sounds that wake or distract you, and the reviews above don't support stronger promises for sleep, focus or ADHD.

## White noise for babies: distance and level

The American Academy of Pediatrics' parent site lists [a white noise device among the sounds that may calm a fussy baby](https://www.healthychildren.org/English/ages-stages/baby/crying-colic/Pages/Calming-A-Fussy-Baby.aspx?ref=soundforgepro.com), alongside a fan's hum and a heartbeat recording. The measurements are what set the limits. [Hugh and colleagues' 2014 study in Pediatrics](https://pubmed.ncbi.nlm.nih.gov/24590753/?ref=soundforgepro.com) found all 14 infant sleep machines louder than 50 dBA at 30 cm at maximum volume, the noise limit recommended for hospital nurseries, and three of them above 85 dBA.

Hong's team drew the practical line: avoid the maximum volume setting, and place any device well outside the crib or at least 30 cm from the child. In their tests, no device passed 85 dBA at 30 cm or at minimum volume.

Duration matters too. In a [2026 survey at a pediatric ENT clinic](https://pubmed.ncbi.nlm.nih.gov/42703389/?ref=soundforgepro.com), 150 of 253 parents (59.3%) used white noise, 90.7% of them overnight, for 9.6 hours a day on average by their own report. A [2024 review by De Jong and colleagues](https://pubmed.ncbi.nlm.nih.gov/38663282/?ref=soundforgepro.com) supports limits on both the maximum level and the duration of white noise devices.

For a baby's room I'd set it up this way: the speaker across the room and never in or on the crib, the level low enough that you can talk over it, and the sleep timer on so it covers falling asleep rather than the whole night. Ask your pediatrician first if your baby was born early or has a hearing concern.

## Keeping it playing through the night

On desktop Chrome, a tab that's making sound keeps running. [Chrome's notes on timer throttling](https://developer.chrome.com/blog/timer-throttling-in-chrome-88?ref=soundforgepro.com) say a page that has made sound in the past 30 seconds gets only minimal throttling, and [Chrome Help on Memory Saver and Energy Saver](https://support.google.com/chrome/answer/12929150?hl=en&ref=soundforgepro.com) lists playing audio among the activities that keep a tab from being deactivated. The sleep timer doesn't depend on page timers anyway: its fade and stop are scheduled on the audio clock when you press Play or change the timer.

iPhone and iPad behave differently. [MDN's page on the audio context state](https://developer.mozilla.org/en-US/docs/Web/API/BaseAudioContext/state?ref=soundforgepro.com) notes that in iOS Safari, switching tabs, minimizing the browser or turning off the screen moves web audio into an interrupted state that has to be resumed. Keep the screen on with the tab in front, or save the WAV loop and play it on repeat in a music player.

The tool won't restart by itself after an interruption, because browsers generally block audio that starts without a tap, click or key press, as [MDN's autoplay guide](https://developer.mozilla.org/en-US/docs/Web/Media/Guides/Autoplay?ref=soundforgepro.com) explains. The status line says what stopped it, and one press of Play brings the noise back. Keep the device on its charger for a long session.

## Limits of this white noise generator

The level is digital, relative to your device's full scale. The same 5% can be faint on a laptop and loud on a Bluetooth speaker, so it isn't a calibrated output or a dB SPL reading.

White and brown are matched for loudness with the K-weighting filter from [ITU-R BS.1770-5](https://www.itu.int/dms%5Fpubrec/itu-r/rec/bs/R-REC-BS.1770-5-202311-I!!PDF-E.pdf?ref=soundforgepro.com), so switching between them doesn't jump in level; left unmatched, white would measure about 8.9 dB louder. Brown here is white noise through a leaky integrator with a gentle cut below about 15 Hz, which keeps it falling at −5.9 dB per octave between 63 Hz and 8 kHz without piling up energy no speaker can play. Softness uses the browser's low-pass filter, which the [Web Audio specification](https://www.w3.org/TR/webaudio-1.1/?ref=soundforgepro.com) defines with a 12 dB per octave roll-off.

Each color is a 60-second loop with separate noise in the left and right channels. In [Kaernbach's listening experiments](https://pubmed.ncbi.nlm.nih.gov/15620225/?ref=soundforgepro.com), untrained listeners could tell repeating noise from fresh noise up to a 20-second cycle, so a 60-second loop sits well past that point. Only white and brown are here; pink is on its own page, and gray, blue and violet aren't offered. For a single steady pitch rather than noise, use the [online tone generator](https://soundforgepro.com/online-tone-generator/).

## White noise generator FAQ

### Is white or brown noise better for sleep?

No trial in the reviews cited here compares them directly. Brown keeps 93% of its power below 250 Hz and has almost no hiss, so it suits low rumble; white covers voices better. Pick the one that hides the noise in your room at the lower setting.

### Is it safe to play white noise all night?

Level and distance decide it. WHO guidance allows adults 40 hours a week at 80 dB, and eight hours a night is 56, so keep the level well below that and the speaker across the room. A 2021 systematic review also warns that continuous noise may affect sleep and hearing, so use the timer if you only need help falling asleep.

### How loud should white noise be?

Only as loud as the sound you want to cover. Published measurements of 20 machines put their loudest settings at a median of about 71 dBA at 1 m, so a machine at full volume is a poor guide. Start at the tool's 5%, raise it until the background fades, and stop there.

### Does brown noise help with focus or ADHD?

There's no direct evidence. A 2024 meta-analysis found a small benefit of white and pink noise on lab attention tasks for people with ADHD or ADHD symptoms (g = 0.249), a small negative effect for people without ADHD, and no brown noise studies at all.

### Can I use white noise for my baby?

The AAP's parent site lists a white noise device among sounds that may calm a fussy baby. Keep it off the crib and at least 30 cm away, never at maximum volume, as Hong and colleagues advise, and use a timer so it doesn't run all night.

### Can I download the noise and play it offline?

Yes. Save WAV loop builds a 16-bit stereo WAV of the current color and softness on your device. A 30-second loop is 5.3 MB at 44.1 kHz or 5.8 MB at 48 kHz; 60 seconds is 10.6 MB or 11.5 MB. The file is built to loop without a click, so set your player to repeat one track.

### Should I use pink noise instead?

Pink falls halfway between white and brown, with equal power in every octave. It has the most positive sleep studies in the 2022 review and is the better signal for speaker and room checks, so use the pink noise generator when either of those is the goal.