How Loud Is a Baby's Cry? Why the Usual Figure Has No Source

Search for how loud a baby's cry is and you will find figures in the region of 110 to 115 dB on parenting sites, hearing-clinic pages and noise blogs. They rarely carry a distance, almost never carry a weighting, and never carry a citation.

We looked for a primary source: a study, a government publication, a standards document, anything that measured infant crying as a sound pressure level under stated conditions. We did not find one. That does not prove no such measurement exists anywhere, but it does mean the figure being repeated has no traceable basis.

There is a serious literature on infant crying, and it is mostly not about loudness. It examines fundamental frequency and its variation, cry duration and rhythm, spectral characteristics, and whether these features carry clinical information about pain or neurological condition.

These are acoustic analyses of recordings. Recording setups optimised for frequency analysis are not calibrated for absolute level, so even where a study reports something about intensity, it typically cannot be read as a sound pressure level at a defined distance from the infant.

A cry is highly variable within seconds: it rises, breaks, pauses for breath and resumes. Peak and average differ substantially, and a figure quoted without saying which is meaningless.

Distance then dominates. A cry heard with your ear next to the infant's mouth and the same cry from across a room differ enormously, and "how loud is a baby's cry" almost never specifies which situation is meant.

Infant cries are concentrated in a frequency region where human hearing is sensitive, and they are acoustically designed — evolutionarily speaking — to be hard to ignore. Their variability adds to that: sound that changes captures attention far more effectively than steady sound at the same level.

So the subjective experience of a cry being overwhelming is real, and it is not evidence for a specific decibel figure.

You can, and the useful output is a comparison rather than an absolute: the difference between the nursery and the next room with the door closed, or how much a white-noise machine raises the background.

Uncalibrated, our meter reports A-weighted digital signal level in dBFS, not sound pressure. After a one-time local calibration it reports an estimated dBA figure. Either way, a brief peaky sound like a cry is exactly the case where a phone microphone is least reliable.

If you are measuring to decide something practical — whether a door seal helps, whether a monitor is turned up further than it needs to be, whether one room carries sound better than another — the comparison answers the question without needing an absolute figure at all.