An open source noise dosimeter for evaluating exposure metrics

Noise dosimeters can be effective tools for measuring individual risk to sound exposure, especially in free-moving environments where sound pressure levels can vary significantly as a function of location and time. Commercial noise dosimeters typically compute noise exposure as an average A-weighted energy over an 8-hour work day (LAeq8hr), and health guidelines suggest a maximum exposure of 85 dB or 90 dB to limit noise-induced hearing injuries. However, recent studies have suggested that conventional noise exposure metrics may not be satisfactory and may not predict susceptibility to synaptic damage in the cochlea nor protect suprathreshold hearing. There is also evidence to suggest that the LAeq may under-predict health risk in complex noise environments, where there is both continuous and impulse noise, and over-predict health risk for long-duration impulse noise. This suggests a need for improved dosimeters with the ability to incorporate new noise metrics. Here we utilize a low-cost open source audio platform to implement standard and experimental noise exposure metrics in an open-source microcontroller programming language. A Bluetooth connection also enables real-time reporting of noise exposures to a smartphone. This system has a significant advantage over using persistent acoustic recordings which have privacy issues and may capture conversations. We envision that this system will be used to further noise exposure research and help evaluate new metrics of auditory health risk. [Work supported by ONR.]