Skip to main content

How Loud Is 70 Decibels?

70 dBA is a normal conversation, a television, a busy office. It is also the last level nobody has to do arithmetic about — and it can still be far too loud for the room you are in.

Seventy decibels is the level at which two different questions stop having the same answer, and almost every argument about noise starts with someone asking one and being given the other.

The hearing question is settled and short. Sounds at or below 70 dBA are unlikely to cause hearing loss even after long exposureNIDCD. No duration to check, no dose to add up, no protection to wear. That is the whole answer.

The liveability question is a different matter, and at 70 dBA the answer is often no.

What 70 decibels sounds like

A normal conversation runs 60–70 dBANIDCD, so 70 is roughly the top of talking. A television at a comfortable volume, a busy office, traffic heard from a pavement, a vacuum cleaner running in the next room — all of them land in that region.

What none of them do is make you raise your voice much. That is the practical marker for this end of the scale: at 70 dBA you can still hold a conversation at arm's length without effort. Fifteen decibels higher, at the 85 dB NIOSH limit, you cannot — you have to lift your voice to be heard three feet awayNIOSH. The gap between "comfortable" and "the level health agencies draw a line at" is narrower than it feels.

Is 70 dB loud?

For your hearing, no, and the guidance says so without qualification. For everything else you use a room for, quite possibly.

The 1974 EPA environmental figures, as NIOSH summarises them
LevelMeasured overWhat it protects
70 dBAaveraged over 24 hoursHearing, for 96% of the population
55 dBAoutdoorsSpeech and annoyance
45 dBAindoorsSpeech and annoyance

Those three figures come from a 1974 EPA report, summarised by NIOSH. It recommended 70 dBA averaged over twenty-four hours as the environmental limit, chosen to protect 96% of the general population from hearing loss, and named 55 dBA outdoors and 45 dBA indoors as the levels above which noise begins to interfere with speech and become an annoyanceNIOSH.

Read the third column and the trap becomes obvious. Seventy is a hearing figure. Forty-five is a living in it figure. They are twenty-five decibels apart, which on this scale is a factor of several hundred in sound energy. A room can be comprehensively unusable — phone calls ruined, concentration gone, sleep broken — while sitting a long way inside the range that will never damage anyone's hearing.

The EPA figures were also never issued as a standard or a regulation, and no US agency enforces themNIOSH. They are a reference point. Local ordinances exist and are enforceable, but they are local, and they are usually written in terms this page cannot generalise about.

Which question do you actually have?

Worth being honest with yourself here, because the answer changes what to do next.

If you are asking about a workshop, a venue, headphones or anything you are in for hours at a time, you are asking about hearing, and 70 dBA is not your problem — the number to worry about is fifteen decibels higher, at the level NIOSH recommends protection aboveNIOSH, and the clock matters more there than the level does.

If you are asking about a neighbour, an office, a flat next to a main road or a child's bedroom, you are asking about annoyance and sleep. Then 70 dBA is very loud, the relevant thresholds are the 45 and 55 figures above, and no amount of hearing-safety literature will support the complaint you are trying to make. It is the wrong body of evidence for the case, and reaching for it is why so many noise complaints go nowhere.

Both are real problems. They just have almost nothing to do with each other, and the same decibel figure answers only one of them at a time.

Why 70 sounds closer to 85 than the numbers suggest

Fifteen decibels reads like a small step. It is not a small step in energy: the decibel scale is logarithmic, and every 3 dB is a doubling of the sound energy reaching your ear — which is exactly why the allowed exposure time halves for every 3 dB above the recommended limitNIOSH. Fifteen decibels is five of those doublings, or about thirty times the energy.

The ear compresses that range so effectively that the difference registers as "a bit louder" rather than "thirty times as much". That compression is a genuinely useful piece of biology and a genuinely misleading one: it is the reason a room can be well past the point where a health agency would ask you to put something in your ears while still feeling merely busy.

At this end of the scale the compression cuts the other way too. Dropping a room from 70 to 60 dBA is a tenfold reduction in energy and will feel, at first, like a modest improvement — and then like a different room after twenty minutes in it.

What 70 decibels costs on the exposure clock

Nothing, and it is worth saying why rather than leaving it implied.

The NIOSH table of exposure limits starts at 85 dBA for eight hours and works upward from thereNIOSH. It has no row for 70, because running its own exchange rate downward that far produces an allowance many times longer than a day. There is no dose to accumulate, no clock to watch, and no arithmetic to do — which is the practical meaning of NIDCD's statement that this level is unlikely to cause hearing loss however long the exposure lastsNIDCD.

So if hearing is your question, a reading around 70 ends the enquiry. If it is not, the number has told you almost nothing, and the next section is the one that matters.

70 dB equivalents, and why they vary

The everyday sources people put at "about 70" cover a genuinely wide spread, which is worth knowing before you plan anything around the number. Published figures come as ranges rather than points because perceived loudness depends on frequency, on how long a sound lasts and on the listenerNIDCD.

Distance is the other reason. The same appliance is a different figure from across the room than from beside it, and most published values do not say from where. A vacuum cleaner is the standard example: NIOSH puts it in a much higher band than the number people repeat, because the published figure is taken at the machine rather than at the far side of the roomNIOSH.

The practical consequence is that a table of everyday sounds tells you which region you are in, not what your room is. The full decibel chart is the region map; the level where you are standing is a different thing, and you can only get it by looking.

Checking a room against 70

This page is attached to an online decibel meter that runs in the browser. Point it at the room, let it settle, and read where the number sits rather than what it touches.

Two things to carry over. The figure is an uncalibrated estimate — smartphone sound apps have been found to scatter by roughly ±10 dB against a reference instrumentKardous & Shaw — and a browser has less to work with than a native app. It is not an IEC 61672 result and must not be used for a legal, occupational, or medical decision.

What an estimate at this precision is genuinely good for is the comparison. The gap between a 45 dBA bedroom and a 70 dBA one is twenty-five decibels, which no plausible error can hide. Reading the room before and after closing a window, or on a quiet night against a loud one, tells you more than one absolute number ever will — and a series of readings taken the same way is worth far more than a single dramatic one.

The other habit worth having: watch the number for a minute rather than a second. A room that hits 70 when a lorry passes is not a 70 dBA room. What matters for both questions on this page is where the level lives, not where it peaks.

Sources

  1. Kardous, Themann, Morata, Lotz — Understanding Noise Exposure Limits: Occupational vs. General Environmental Noise, NIOSH Science Blog, 2016. https://www.cdc.gov/niosh/bulletin/2016/noise.html
  2. NIOSH — Noise-Induced Hearing Loss. https://www.cdc.gov/niosh/noise/about/noise.html
  3. NIDCD — Noise-Induced Hearing Loss. https://www.nidcd.nih.gov/health/noise-induced-hearing-loss
  4. Kardous & Shaw — Evaluation of smartphone sound measurement applications. https://pmc.ncbi.nlm.nih.gov/articles/PMC4545478/