Setting Up Your Remote Work Sound Environment

Two rooms, two different problems

One home office sits under a constant ventilation hum. Another, at the same average level on the same meter, has a dog that barks eleven times a day. The first room is a masking problem: a steady sound can cover another steady sound without ever drawing attention. The second is an onset problem, where the average level says nothing useful and covering the peak is what matters. The two rooms need opposite setups, and the meter cannot tell them apart.

Buying hardware before establishing which room you are in is how people end up with expensive noise cancelling that does nothing about the thing that actually breaks their concentration.

So the order here runs backwards from the usual advice. Characterise the room, close the paths that are cheap to close, then choose headphones for what is left, and add sound last — because added sound is the only item on that list whose job is defined entirely by what the previous three leave behind.

Two categories, not a decibel reading

The useful first measurement needs no instrument and is not a number. Sit in your working position at a normal working hour, do nothing else for five minutes, and write down every sound you can hear. Then sort the list into two columns: sounds that are on continuously, and sounds that start and stop.

The split matters because the two have different remedies and only one of them is a good match for masking. A continuous sound is a candidate for habituation and for masking, since a steady background can cover another steady background without ever drawing attention. An intermittent sound is the harder case: it is the onset that costs you, masking has to be loud enough to cover the peak rather than the average, and if the intermittent sound carries speech then level is not even the relevant property. Experimental work on office noise is specific that the disruption from speech is driven by how intelligible it is, not how loud.

A smartphone sound-level app can add a rough figure if you want one, but treat it as uncalibrated and do not build decisions on the digits. The categorisation is what changes your plan.

Four paths, and they arrive in order

Sound gets into a room by more routes than the one people think about, and the routes have different fixes. The reason this matters is that building acoustics distinguishes a partition's laboratory rating from what a real room achieves: the standard single-number rating for a wall or door, sound transmission class, is derived from measurements in a laboratory across the one-third-octave bands from 125 to 4000 hertz, and the corresponding field quantity is called apparent sound transmission class precisely because real installations leak around the partition rather than through it.

A section through a home office with a listener seated at the centre wearing headphones. Four numbered paths reach the listener. Path one comes from the street through the glazing, rated by ASTM E90 as sound transmission class. Path two comes from the hallway, round the door and through the gap at the threshold, with a note that a gap bypasses the door's rating entirely. Path three comes through the structure as footsteps from the floor above. Path four originates in the room itself: your own fan or boiler, plus every hard-surface reflection of the first three. A closing note states that paths one to four set the level arriving at the chair, and only then does it matter what the headphones attenuate, and only then is there a gap left for added sound to cover. The room Listener Headphones and whatever you add Street Traffic, works 1. Through the glazing Rated by ASTM E90, as sound transmission class Hallway Voices, kitchen 2. Round the door, through the threshold gap A gap bypasses the door’s rating entirely 3. Through the structure: footsteps from above 4. In the room Your own fan or boiler, plus every hard-surface reflection of 1 to 3 Paths 1 to 4 set the level arriving at the chair. Only then does it matter what the headphones attenuate, and only then is there a gap left for added sound to cover.
The four paths into a home office, in the order they act. No attenuation figures are drawn: a partition's laboratory rating does not predict a particular installation, which is why the field quantity is called apparent sound transmission class. The ordering is the point — headphones and added sound act on what these four leave.

That distinction is the practical lesson in one sentence: a gap defeats a rating. A solid door with a centimetre of air under it performs far worse than its rating suggests, and no amount of upgrading the door itself addresses that, because the sound is not going through the door. Sealing the perimeter of an existing door is the cheapest intervention available in most home offices, and it is cheap for exactly this reason — you are not improving a partition, you are removing a bypass.

Windows are the next path, and here the honest answer is less encouraging. Curtains and soft furnishings absorb sound inside the room, which reduces reflections and makes the space less reverberant, but they do very little about transmission from outside. If street noise is your dominant problem, heavy curtains will change the character of the room more than the level arriving at your desk, and the real fix is glazing, which is a building project rather than a purchase.

Structure-borne sound — footsteps from the floor above, a washing machine through a shared wall — travels in the building rather than the air, and nothing you place inside the room intercepts it. This is the path to identify early, because if it is your main problem the entire sequence below is a partial measure and relocating within the home will do more than any equipment.

Last is the room itself: your own fan, a drive, a boiler, plus every hard surface re-reflecting the first three paths. Rugs, shelving and upholstery act here and only here. They are worth doing, and they are not soundproofing.

Then the headphones

With the paths characterised, the headphone choice becomes a narrow question instead of an open one. Closed-back over-ear designs attenuate across a broad range passively, which is useful against speech and other mid and high-frequency intrusions. Active cancellation works best on steady low-frequency sound, which is ventilation and traffic rumble, and is weaker against the sudden and the high-pitched. If your list from the five-minute exercise is mostly voices and doors, cancellation is the wrong lever and passive isolation plus a modest added sound is the right one. If it is mostly a constant mechanical drone, cancellation is addressing your actual problem.

The attenuation figures printed on boxes are not comparable between designs or verifiable at your desk, so this article does not reproduce any. The decision does not need them; it needs to know which column your noise sits in.

Where added sound fits, and how much

Added sound is the last layer because its only job is to cover what survives the first three. That makes it smaller than most setups treat it, and it also means the level you need is lower if the earlier steps were done — which is the real argument for doing them.

Two limits bound the choice. The upper one is hearing safety: the National Institute on Deafness and Other Communication Disorders reports that sound at or below 70 A-weighted decibels is unlikely to cause hearing loss even after long exposure, while long or repeated exposure at or above 85 dBA can cause it, with normal conversation at 60 to 70 dBA. For reference on the other side, published target values for installed office masking put the useful band at roughly 40 to 45 decibels, and recommend against exceeding 45 because people start raising their voices. An all-day personal level belongs nearer the second figure than the first.

The lower limit is that the sound is not free. A meta-analysis of sixty-five reading experiments found background noise, speech and music all carry a small but reliable cost, with intelligible speech and lyrics worst. The justification for adding sound is never that it helps; it is that the thing it covers is worse. So the rule is to raise it until the intrusion stops pulling at you, and then to stop, which in a room where the cheap paths have been closed is quieter than people expect.

For what to play: brown noise has most of its energy low and covers rumble at a low level; white noise is flat and therefore has energy wherever speech is; pink sits between them and is the easiest of the three to sit under all day. Running a mostly brown mix with a thin white layer on top is a reasonable compromise when the intrusion spans both ends. Avoid anything with recognisable content, since the mechanism is about putting no information in the background.

Shared homes

In a shared home the asymmetry is the problem: your working hours are a constraint on other people's behaviour, and no acoustic measure substitutes for saying which hours those are. The one technical point worth adding is directional: a closed-back headset keeps call audio from reaching the rest of the home, and a close-talking microphone keeps less of your voice travelling the other way. Both of those are about the sound you generate, which the rest of this article does not address and which is often the half that causes friction.

References

Frequently Asked Questions

What is the most cost-effective way to improve home office acoustics?

Sealing the perimeter and threshold of an existing door, because a gap bypasses the door's rating altogether and closing it is not an upgrade to the partition but the removal of a shortcut around it. It is also the only one of the four transmission paths that is both cheap and genuinely a transmission fix rather than an in-room absorption fix.

Are active noise-cancelling headphones needed for remote work?

It depends which column your noise falls into. Cancellation is strongest against steady low-frequency sound such as ventilation and traffic rumble, and weaker against sudden or high-pitched sounds including voices. If the five-minute listening exercise produces a list of mostly intermittent and speech-like sounds, passive isolation plus a modest added sound addresses them better.

Do heavy curtains block outside noise?

Barely. Soft furnishings act on reflections inside the room, which is a different quantity from transmission through the window, so they make a space less reverberant without much changing the level arriving from the street. Where external noise dominates, the effective fix is glazing, which is a building change rather than a purchase.

Can added sound replace a quiet room?

No, and the limit is structural rather than a matter of degree. Masking replaces an irregular distraction with a steady one, which is a trade; if covering the intrusion requires a level you would not otherwise choose, the sound has stopped being the right tool. Structure-borne noise such as footsteps from above is the clearest case, since nothing placed inside the room intercepts it.

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.