Most articles about sound masking are written by companies that only sell installed systems. So they all reach the same conclusion, which is that you need an installed system.
We install them, and we still think that’s dishonest. There are real situations where a single desktop device solves your problem for a couple hundred dollars, and telling you otherwise would cost you money you didn’t need to spend. There are also situations where a device will waste your money entirely, and the difference between the two comes down to physics you can reason about in about five minutes.
Here’s how to tell which one you’re in.
What a masking device actually does
Sound masking adds a low, engineered background sound tuned to the frequency range your ear uses to turn sound into words — specifically the consonant range, where intelligibility actually lives. Nothing gets cancelled and nothing gets covered. You raise the room’s background level just enough that speech past a certain distance stops being decipherable. You can still tell someone is talking. You just can’t make out the words.
Acousticians call the distance where that happens the radius of distraction. In a quiet room it can run 30 to 40 feet.
A desktop device does the same thing as an installed system. The difference is not quality. The difference is coverage, and it’s governed by one unavoidable rule: sound from a single point source drops off sharply as you move away from it — roughly 6 dB every time you double the distance. That’s why one device on a credenza can protect a doorway or a desk, and why the same device does nothing useful 30 feet away. You cannot fix that by turning it up. Turning it up just makes it irritating where you’re sitting, without extending the useful radius much.
An installed system solves the coverage problem by not being a point source. Emitters sit on a grid above the ceiling and fire upward into the plenum, so the sound arrives diffuse and even — no hot spots, no dead zones, nothing you can point at.
That single distinction decides everything below.
Four situations where a desktop device is the right call
One private office where sensitive conversations happen. An HR manager, a practice administrator, a principal, a therapist. One door, one hallway, one adjacent desk that shouldn’t hear anything. A device placed between the conversation and the leak point — usually just inside the door or on a credenza against the shared wall — does the job.
A reception or intake desk with seating behind it. The classic exposure: a patient or client in a waiting chair can follow the check-in conversation word for word. One device near the desk, aimed at the seating, is often genuinely sufficient. This is also the single most common real HIPAA exposure we find in small practices.
A room you don’t control. A leased suite where you can’t get above the ceiling, a hard-lid drywall ceiling with no plenum access, a space you’ll be out of in eighteen months, or a building where facilities won’t approve cabling. A device needs a power outlet and nothing else.
A single interview, deposition or exam room. One room, one purpose, a clear leak path. Courthouses and law offices ask us about this constantly, and for one room the honest answer is often a device.
The pattern: one room, one leak path, and no requirement to prove anything to anyone.
Four situations where a device will disappoint you
Any open plan. This is the one people get wrong most often. Open plan is precisely the case a point source cannot serve — you’d need a device on every desk, they’d each be audible as a separate source, and the level would be wrong everywhere except within a few feet of each one. Twelve desktop units is not a system, it’s twelve problems.
More than about two rooms. The math turns against you fast. Once you’re buying a device per room, you’re approaching installed-system pricing without the uniformity, the zone control, the scheduling, or the ability to tune it.
When someone is asking you for proof. If the conversation involves a compliance officer, a HIPAA risk assessment, a client security questionnaire, or an architect writing a privacy class into a spec, a consumer device gives you nothing to show. There’s no commissioning, no measurement, no documentation of what the room actually achieves.
When it has to be uniform. People notice sound that changes as they walk. A room where the masking is obvious at one end and absent at the other is more distracting than no masking at all. Uniformity is the thing installed systems buy you, and it’s the thing devices cannot.
Side by side
| Desktop device | Installed system | |
|---|---|---|
| Realistic coverage | A work area, not a room | Whole floors, uniformly |
| Open plan | No | Yes — this is the primary use case |
| Uniform level | No, falls off with distance | Yes, diffuse from a ceiling grid |
| Zone control & scheduling | No | Yes, per area |
| Tuned to your room | No, fixed spectrum | Yes, commissioned and measured |
| Documentation for compliance | None | Measured results at handoff |
| Paging or background music | No | Yes, same emitters |
| Typical cost | A few hundred dollars | From a few thousand installed |
| Install time | Unbox it | One day for a small suite |
If you’re buying a device, what to look for
Skip anything marketed as a white noise machine for sleep. Those produce untuned hiss, which is loud where you’re sitting and doesn’t match the speech range — you pay in comfort and get little privacy back.
What matters:
- A speech-shaped spectrum, not flat white noise or nature sounds. If the product literature never mentions speech privacy, it’s a sleep machine.
- Fine level control. You’re targeting a specific background level, not “louder.”
- No loop. Anything with an audible repeat becomes maddening within a week.
- Continuous operation. It should run all day without a timer shutting it off.
- No music, no rainfall, no waves. Anything with recognisable content competes for attention instead of reducing competition.
One placement note worth more than the product choice: put the device between the conversation and the person who shouldn’t hear it, not next to the person talking. You’re raising the background level at the listener’s ear. Most people get this backwards and then conclude masking doesn’t work.
What it costs, either way
A device is a few hundred dollars and you can order it this afternoon. We don’t sell them and we don’t take a cut.
For installed systems, we publish honest starting brackets on our sound masking installation page — a small professional suite is often addressed for a few thousand dollars installed, and the number moves with square footage, ceiling type, how many zones you want to control separately, and whether you add paging or music. We don’t quote before we’ve measured, but you’ll leave an assessment visit with a budget range you can take to a board or a finance committee.
Deciding in five minutes
Answer these four:
- How many rooms need protecting? More than two, go installed.
- Is any of them open plan? Yes, go installed.
- Does anyone need documentation or a privacy class? Yes, go installed.
- Can you get above the ceiling, and will you be in this space in three years? No to either, a device may be the smarter spend.
One room, one leak path, nothing to prove, no ceiling access: buy a device. Anything else, an installed system is cheaper per square foot and it’s the only option that actually works in open plan.
Common questions
Does sound masking really work?
Yes, and it’s measurable — but it does one specific thing. It makes speech unintelligible past a short distance. It does not make a room quieter, it does not stop footfall or HVAC rumble, and it does not fix a reverberant room. Those are absorption and isolation problems. If your complaint is “I can hear what they’re saying,” masking is the right tool. If it’s “this room is loud and echoey,” you need acoustic treatment, and possibly both.
How does it work, physically?
It raises the room’s noise floor with sound engineered to sit in the same frequency band as speech consonants. Intelligibility depends on the gap between the speech level and the background level. Narrow that gap and the words stop resolving, even though the voice is still audible.
Will a device stop someone eavesdropping deliberately?
It raises the effort considerably at normal conversational distance, and a device at the door helps meaningfully with casual overhearing in a corridor. It is not a countermeasure against someone determined and close, and nobody should sell it to you as one. Deliberate-threat environments need isolation and procedure, not just masking.
Will my staff hate it?
Untuned consumer hiss, sometimes — legitimately. A properly commissioned installed system, almost never; most people stop noticing inside a day or two. The failure mode is always level and spectrum, not the concept.
Do these work in a home office?
For one room with one leak path, the logic above holds identically. We don’t do residential installs, but the device guidance applies unchanged.
When you want the installed version
We design and install Biamp Cambridge systems across New Jersey, eastern Pennsylvania and Delaware, and we measure at both ends — before, so we know what the room is doing, and after, so you know what you bought. Recent work includes Princeton University Press, Gallagher Marine Systems and M&T Bank.
If you’re past the one-room stage, request a free acoustic assessment and we’ll tell you honestly whether you need us.