A thousand people a month search for this. Of the seventeen machines you can currently buy from the brands here, exactly one has a published decibel figure.
Disclosure — some links on this page go to Amazon, and we earn a commission if you buy through them. It costs you nothing extra and it does not change what we recommend or in what order. The full disclosure.
Almost no manufacturer publishes a noise figure. Of ten current product pages we checked on 2026-09-14, only ECOVACS states one for a model sold here: 65 dBA for the DEEBOT T30S Combo cleaning in standard mode and 78 dBA for its dock emptying. Everyone else offers adjectives. Since the dock is the loud part, the checkable proxy is how rarely it runs.
Noise is one of the three things people ask about before buying a robot vacuum, alongside whether it will eat the rug fringe and whether the dog will hate it. It is also the specification the industry has quietly agreed not to publish. That combination — high demand, no data — is worth a page on its own, because the shopping advice written around it is almost entirely invented.
What ten manufacturer pages actually say
We went to the product page for every brand we cover and looked for a decibel figure. Not a review, not a forum measurement — the maker's own published number. This is the result, checked on 2026-09-14.
Published noise figures, by model
Model
What the maker publishes
ECOVACS DEEBOT T30S Combo
65 dBA sweeping and 65 dBA sweeping with mop, both in standard mode; 78 dBA dust collection at the station
Roborock Saros 10R
No figure
Roborock Qrevo Curv series (Curv, CurvX, Curv S5X)
No figure
Roborock Q7 M5
No figure
Dreame X50 Ultra
No figure. States "Lower Noise" and that noise levels are kept "impressively low"
eufy X10 Pro Omni
No figure
eufy Omni C28
No figure, and no quietness claim of any kind
iRobot Roomba Max 715 Pet
No figure, and no quietness claim of any kind
iRobot Roomba Combo j9+
No figure. States the docking base is "our quietest yet"
Shark PowerDetect 2-in-1
No figure
One published figure out of ten pages. Where a maker says nothing we print nothing — we are not substituting a reviewer's meter reading or an estimate of our own.
Two of those entries are worth reading twice. iRobot says the j9+ dock is its quietest yet, which is a comparison against unnamed previous iRobot docks and tells you nothing about how it compares with anyone else's. Dreame makes two separate quietness claims about the X50 Ultra and attaches no number to either. Both are the same move: the benefit without the measurement.
The makers can publish this. They do it elsewhere
The most revealing part of the research is that the silence is selective. Roborock publishes a decibel figure for the Saros 10 — 52 dB(A) in quiet mopping and 72 dB(A) for dock self-emptying — on its own product page. It is a model we do not cover, and Roborock publishes nothing equivalent for the Saros 10R, the Q7 M5 or any Qrevo Curv.
ECOVACS does the same thing in the other direction. Its quiet-vacuums category page states that a DEEBOT can be "as low as 65dB", and gives 65 dB sweeping and 68 dB sweeping-and-mopping as brand-level claims not attached to any particular machine — then names one model, the N30 PRO OMNI, at "from 68dB during sweeping and mopping". A brand-level range is not a specification for the robot you are about to buy.
So this is not a measurement problem. These companies own anechoic chambers and publish dBA figures in other markets and on other models. The figure is simply absent from the pages where an American buyer would look for it.
The one machine with numbers, and what they mean
The ECOVACS T30S Combo publishes three figures, and the gap between them is the single most useful thing on this page: 65 dBA while the robot cleans, 78 dBA while the station empties it.
Read the mode label carefully too. ECOVACS attaches its 65 dBA to standard mode, not quiet mode, and publishes no quiet-mode figure. So the one machine here with a number does not give you the number for its quietest setting — which is the setting anybody shopping on this page would actually use.
The dock is the loud part, and that is the lever you have
Everything above points at the same conclusion. The robot itself, on carpet, in a normal mode, is a background appliance. The self-emptying burst is the event people actually complain about: it is short, it is sudden, it is the loudest thing the machine does, and on the only model with published figures it is 13 dBA above everything else.
You cannot make that burst quieter. You can make it happen less often, and how often it happens is published — it is the one noise-adjacent number that every maker does state, just under a different name.
Stated interval between self-empty cycles, longest first
Model
Maker's stated interval
Bursts per year, roughly
iRobot Roomba Max 715 Pet
90 days
4
iRobot Roomba 105 Vac
Up to 75 days
5
Roborock Qrevo Curv S5X
Ten weeks
5
Roborock Q7 M5
Seven to nine weeks
6 to 7
iRobot Roomba Combo j9+
Up to 60 days
6
Shark PowerDetect 2-in-1
60 days
6
These are bag-change intervals as each maker states them, for an unspecified average household. They are not a promise for a house with two dogs in it — see self-emptying vs a manual bin for what the intervals mean in practice.
A word of caution on reading that table as a noise ranking. The interval is how often the bag is emptied by you, and most of these robots return to the dock and transfer their on-board bin far more often than that — typically after every run. The long intervals describe how rarely the dock's own bag fills, not how rarely the loud transfer happens. It is still the right direction to reason in, because a dock designed around a large sealed bag is a dock designed to be visited less. But it is a proxy, and we would rather say so than dress it up as a measurement.
Why "quiet mode" is not the answer you want
Every machine here has a quiet or eco setting, and every one of them solves the noise problem by reducing suction. That is a real trade, not a free one: the mode that makes the robot acceptable during a phone call is also the mode least likely to get grit out of carpet pile.
There is a second catch, and it is one this site has flagged before. Published runtime figures are quoted in the quietest mode on the easiest surface — the 291-minute and 220-minute headline numbers are quiet-mode figures on hard floor. So quiet mode is already baked into the runtime you were sold, and running in a normal mode costs you both the quiet and a large part of the battery life. The detail is in battery life and runtime compared.
What actually makes a robot quiet in your house
Given that the industry will not give you a number, here is what genuinely moves the needle, in rough order of effect.
Run it when you are out. This is not a cop-out answer, it is the whole answer. Every one of these machines schedules, and a robot running at 10am in an empty house is a 0 dB robot from your point of view. Nothing on any spec sheet beats it.
Put the dock somewhere with a door. A utility room, a hallway cupboard, under the stairs. The 78 dBA burst through a closed door is a different event. This is the single highest-value physical change you can make and it costs nothing.
Keep it off hard floor at night if you can. A rigid chassis on tile transmits more than the same machine on carpet, and the wheels and the brush contribute as much as the fan on a hard surface.
Use zones rather than quiet mode. Excluding the room you are sitting in keeps full suction everywhere else. Most people reach for the mode switch when the map would serve them better — see mapping tips.
Expect the hot-air drying cycle. Docks that dry the mop pad run a fan for hours after the clean. It is much quieter than the emptying burst but it is not silent, and nobody publishes a figure for it either.
How we picked
Criteria first, products second. These are the things we ranked on, stated before anything is recommended.
A published decibel figure, from the manufacturer. Not a reviewer's meter reading, not our estimate. We checked ten current product pages and print exactly what each one says.
How often the loudest event happens. On the one machine that publishes both numbers, the dock emptying is 13 dBA above the robot cleaning. A dock that empties every ten weeks interrupts you a sixth as often as one that empties weekly.
Whether the maker states a mode with the figure. A decibel number without a mode attached is close to meaningless — the same machine on maximum suction and on quiet is a completely different noise.
Published specifications generally. A maker that states a noise figure at all is a maker willing to be held to it, and that habit tends to run across the whole spec sheet.
What we are NOT ranking on: our own impressions of loudness. We have not run these machines and we do not own a sound meter. Every ranking below is built on what is published or on arithmetic you can check.
What we did not do: test any of these machines. We do not own one, we do not have a lab, and nothing on this page is written as though we did. The full method.
Quick picks
Ranked on the published specifications below. Tap a row to jump to the full write-up.
#
Product
Best for
Suction
Price
1
ECOVACS DEEBOT T30S ComboShips with a cordless stick vacuum docked alongside the robot, which is the only honest answer to stairs anyone sells.
iRobot Roomba Max 715 PetDual rubber anti-tangle brushes and a 90-day allergen-sealed dock — and the one place iRobot has printed an actual suction number instead of a multiple.
iRobot Roomba 105 VaciRobot's entry machine, and the first cheap one with real LiDAR — but it keeps the bristle brush, which is the wrong brush for a shedding household.
Dual rubber anti-tangle brushes and a 90-day allergen-sealed dock — and the one place iRobot has printed an actual suction number instead of a multiple.
It is the most capable machine in this registry on almost every other page — the highest published climb, 20,000 Pa, a complete spec sheet. On this page it is the clearest example of the problem. Dreame's own product page says the ProLeap system delivers "Lower Noise" and that its shock absorption keeps "noise levels impressively low", and then publishes no figure at all: not for the robot, not for the station, not in any mode. Two quietness claims and zero numbers, at flagship money. Buy it for the climb height, which is published and checkable. Do not buy it because it is described as quiet.
What would have to change for us to rank this properly
One line on each product page: the A-weighted sound level of the robot in each cleaning mode, and of the station during dust collection, measured at a stated distance. ECOVACS does it for the T30S Combo. Roborock does it for the Saros 10. Neither does it consistently, and nobody else does it at all.
Until that changes, a page claiming to rank the quietest robot vacuums by decibel is either quoting one manufacturer, quoting somebody's phone app, or making it up. We would rather publish the gap. If you find a maker's published figure we have missed, send it to us with the link and we will add it with the source and the date. Corrections here get dated rather than silently edited — the policy is on how we pick.
If quiet matters to you because the machine will be running in one room while you live in it, the more useful page is probably best robot vacuums for a small apartment or studio, where the dock footprint and the scheduling question get the same treatment.
Questions people actually ask
+ What is the quietest robot vacuum?
Nobody can answer that honestly from published data, because only one of the models covered here has a published figure. The ECOVACS DEEBOT T30S Combo states 65 dBA cleaning in standard mode and 78 dBA at the dock. Every other maker we checked publishes no decibel figure at all, so any ranking of the rest is somebody's estimate rather than a specification.
+ How many decibels is a robot vacuum?
The only manufacturer figures we could verify for models sold here are ECOVACS' 65 dBA for the T30S Combo cleaning in standard mode and 78 dBA for its station emptying, plus Roborock's 52 dB(A) quiet mopping and 72 dB(A) dock self-emptying for the Saros 10, which is a model we do not cover. Treat any other number you see as unsourced unless it names the maker's own page.
+ Is a robot vacuum loud enough to wake a baby?
The cleaning itself is around conversation level on the one machine with a published figure. The self-empty burst is the problem: 78 dBA, sudden, and lasting several seconds. Put the dock behind a door and schedule runs for when the room is empty, which solves it far more reliably than choosing a different model on unpublished claims.
+ Why is the dock louder than the robot?
It has to move the entire contents of the robot's bin up a tube and into a bag in a few seconds, which takes far more airflow than cleaning does. On the one model publishing both figures the gap is 13 dBA, and because decibels are logarithmic that is closer to two and a half times as loud rather than twenty percent.
+ Does quiet mode actually help?
Yes, and it costs you suction, because that is how it works. Worth knowing: published runtime figures are already quoted in the quietest mode on the easiest surface, so quiet mode is baked into the battery life you were sold rather than being a bonus on top of it.
+ Which brands publish noise figures at all?
ECOVACS, for the T30S Combo, on its product page spec table. Roborock, for the Saros 10, but not for the Saros 10R, the Q7 M5 or any Qrevo Curv. Dreame, eufy, iRobot and Shark publish no decibel figure for any model we cover — though Dreame and iRobot both make quietness claims in words.
