Knowledge of the auditory and non-auditory effects of noise has increased dramatically over the past
decade, but indoor noise exposure measurement methods have not advanced appreciably, despite the
introduction of applicable new technologies. This study evaluated various conventional and smart
devices for exposure assessment in the National Children’s Study. Three devices were tested: a sound
level meter (SLM), a dosimeter, and a smart device with a noise measurement application installed.
Instrument performance was evaluated in a series of semi-controlled tests in office environments over
96-h periods, followed by measurements made continuously in two rooms (a child’s bedroom and a most
used room) in nine participating homes over a 7-day period with subsequent computation of a range of
noise metrics. The SLMs and dosimeters yielded similar A-weighted average noise levels. Levels measured
by the smart devices often differed substantially (showing both positive and negative bias, depending on
the metric) from those measured via SLM and dosimeter, and demonstrated attenuation in some frequency
bands in spectral analysis compared to SLM results. Virtually all measurements exceeded the
Environmental Protection Agency’s 45 dBA day–night limit for indoor residential exposures. The measurement
protocol developed here can be employed in homes, demonstrates the possibility of measuring
long-term noise exposures in homes with technologies beyond traditional SLMs, and highlights potential
pitfalls associated with measurements made by smart devices