Combining Ambient Sounds with Breathing Rhythms

Two Tabs, Breathing in Front

Open the mixer at /en/mixer in one tab and the breathing panel at /en/breathing in another, then bring the breathing tab to the front and leave it there for the session. The two tools are not connected and do not share a page, and the breathing panel produces no sound of its own — no tone at a phase boundary, no chime when the session ends — so running both means running two tabs, and which one sits in front has a documented consequence.

Browsers throttle timers in tabs that are not in front. The MDN reference for setTimeout records the specifics: Firefox on the desktop enforces a minimum delay of one second in an inactive tab, Firefox for Android fifteen minutes, and Chrome drops to checking timers once a minute for pages that have been invisible for more than five minutes and silent for more than thirty seconds. Firefox waives the throttling for a page holding a Web Audio AudioContext, and Chrome keeps a page in its lightest tier if it has made sound recently.

Those waivers apply to the sound tab, not the breathing tab. Audio keeps playing in the background; timer callbacks are what get delayed. So the tab to leave in front is the breathing tab, and the mixer can sit behind it. Reverse that and the phases stretch unpredictably, which defeats the only thing a timer is for.

Two Interventions, Not One

The usual framing of this topic assumes a single combined thing with combined effects. It is not one thing. It is two interventions that happen to be running at the same time, each with its own evidence, and the useful question is what each one contributes rather than what the pair does together.

Two stacked tracks on a shared time axis running from zero to thirty-eight seconds, which is two complete 4-7-8 cycles at one second per count. The upper track is the breathing phase, drawn as a row of labelled segments: a four-second inhale, a seven-second hold in a fainter weight, an eight-second exhale, and then the same three again. Five vertical lines mark the phase boundaries and run down through both tracks. The lower track is the noise level, drawn as a single horizontal line with a light wash beneath it, running at one constant height from the start of the axis to the end and crossing every phase boundary without deflecting. It is labelled set once then constant, and annotated that there is no change at any phase boundary. The upper track is annotated as changing six times. In 4 Hold 7 Out 8 In 4 Hold 7 Out 8 Breathing phase Changes 6 times No change at any phase boundary Noise level Set once, then constant Level Silence 0 10 20 30 38 Time (seconds) Two complete 4-7-8 cycles
Two 4-7-8 cycles against the sound playing alongside them. The breathing phase changes six times; the level does not change once. The flat line is not a simplification — no amplitude modulation is tied to the breath, and on this site the two tools are on separate pages. The level drawn is a slider position rather than a measured decibel value, which is why no number is printed on it.

The figure is the situation drawn out. Over two cycles the breathing phase changes six times while the sound does not change at all. No amplitude modulation is tied to the breath, no level rises with the inhale, nothing in the player is listening to the timer. Whatever the combination gives you, it is not coupling.

What the Sound Half Rests On

The evidence for background sound is narrower than the genre of advice around it suggests, and it mostly points the other way.

A 2018 Bayesian meta-analysis of sixty-five studies on auditory distraction during reading found that background noise, speech and music all carried a small but reliable cost to reading performance, with intelligible speech and lyrical music the most disruptive. A 2012 experiment in the Journal of Consumer Research found the opposite sign for a different task: a moderate level of ambient noise improved performance on creative idea generation, while a loud level did not. A 2021 systematic review of noise as a sleep aid graded the evidence that continuous noise improves sleep as very low quality, and noted that some studies find it disrupts sleep instead.

What the mixer offers for the purpose is five generated layers — white, pink, brown, green and grey noise, each with its own slider and on-off switch — rather than recordings. All five are steady by construction, which is the property that matters here: a recording of rain or waves varies from moment to moment, and a sound that changes is a sound that can catch your attention at the wrong point in a cycle.

Read together, those give a defensible use and an indefensible one. Sound is not a general focus aid, and nothing establishes that it beats quiet. Where it earns its place is masking: a steady sound covering an intermittent one — a conversation, a door, traffic — replaces a distraction that grabs attention with one that does not. If the room is already quiet, that argument has nothing to work on.

What the Breathing Half Rests On

The breathing side has a clearer mechanism and a shorter list of demonstrated outcomes.

Slowing the breath raises heart rate variability while it is happening, and the most direct comparison of patterns is recent. In a 2025 experiment, eighty-four people breathed four ways in a single session — square breathing, 4-7-8 breathing, and six breaths a minute at two inhale-to-exhale ratios — and six a minute raised heart rate variability more than either named pattern, while no condition meaningfully moved blood pressure or mood.

Two reviews explain why six recurs. A 2022 review locates a resonance near 0.1 Hz, which is six breaths a minute, and a 2017 physiological review restricted to rates between four and ten a minute places what it calls autonomically optimised respiration in the band of six to ten, achieved by breathing more deeply rather than by holding still. The article on 4-7-8 breathing on this site goes through that physiology pattern by pattern; for this page the relevant point is that none of these studies compared breathing with sound against breathing in quiet.

Nobody Has Tested the Pair

No published study takes paced breathing with ambient sound and compares it against paced breathing in quiet. That is the honest centre of this topic, and it means every claim about the combination is extrapolation from the two halves.

Extrapolating is not automatically wrong, but it is worth noticing which direction the halves point. The breathing effect is physiological and does not obviously depend on the acoustic environment. The sound effect, where it exists, is about covering an interruption — and an interruption during a breathing session costs you the cycle you were counting, which is a real thing a steady sound can protect. That is a modest and plausible case for the combination: the sound is not improving the breathing, it is reducing the chance that something stops it.

What has no support is the stronger framing — that sound and breath reinforce each other, occupy complementary senses, or add up to more than their parts. Those are claims about an interaction, and an interaction is exactly what nobody has measured.

Setting the Level

Advice to set the volume to a specific percentage is the most common unsupported claim in this area, and it is unsupported twice over.

First, a percentage is not a level. The mixer’s sliders run from 0 to 100 and set a gain multiplier on the generated noise; what reaches your ear also depends on the system volume, the headphones and their efficiency. The same slider position is a different sound level on two different setups, so no percentage can be a recommendation.

Second, no source establishes an optimal level for this. The one thing that is well established is the safety ceiling. The National Institute on Deafness and Other Communication Disorders puts sounds at or below 70 A-weighted decibels as unlikely to cause hearing loss even after long exposure, states that long or repeated exposure at or above 85 dBA can cause it, and gives normal conversation as 60 to 70 dBA. A sound you intend to leave running for an hour belongs well below those figures.

Which leaves a usable rule that does not pretend to be a measurement: set the level by what it has to cover. Raise it until the sound you are trying to mask stops pulling your attention, then stop. If there is nothing to mask, the honest answer is that you do not need it.

References

Frequently Asked Questions

Does the sound follow the breathing rhythm?

No. Nothing in the player responds to the breathing timer: there is no amplitude modulation tied to the breath, no tone at a phase boundary and no sound when a session ends. On this site the two tools are on separate pages — the noise layers at /en/mixer, the breathing panel at /en/breathing — so combining them means two tabs running side by side.

Which ambient sound works best with paced breathing?

Nothing establishes an answer, because no study compares sounds in this context. The sound evidence that does exist is about masking: a steady sound covering an intermittent one. On that basis any steady broadband noise will do, and the choice between white, pink and brown is a matter of which you would rather listen to.

How loud should the sound be?

No source establishes an optimal level, and a slider percentage is not a level anyway — the same position is a different sound on different headphones. What is established is the ceiling: the NIDCD puts sounds at or below 70 dBA as unlikely to damage hearing even after long exposure and says 85 dBA and above can with long or repeated exposure. Set it just high enough to cover whatever you are masking, and no higher.

Is the combination of sound and breathing better than either alone?

Unknown. No published study compares paced breathing with ambient sound against paced breathing in quiet, so claims that the two reinforce each other describe an interaction nobody has measured. The modest defensible case is different: a steady sound reduces the chance that an interruption costs you the cycle you were counting.

WhiteNoise.top

WhiteNoise.top builds free browser-based noise generators, mixers and breathing timers. Our articles explain the acoustics behind those tools and cite published research where it exists.