# Evaluating the Accuracy of Android Applications in Monitoring Environmental Noise Levels

## TL;DR

This study tested Android apps for measuring noise levels and found some are accurate at low levels but less so at higher ones.

## Contribution

The study provides empirical evaluation of Android apps' accuracy for environmental noise measurement in occupational health contexts.

## Key findings

- The control meter showed highest accuracy (R2 = 0.99) while top-performing apps had R2 = 0.98.
- App accuracy declined at higher noise levels, with Sound Meter (ABC Apps) showing R2 = 0.85.
- User ratings did not consistently correlate with app performance.

## Abstract

Objective

This study aimed to evaluate the accuracy of Android applications (apps) in measuring environmental noise levels, focusing on their potential use for occupational health assessments.

Methods

The top 10, highly rated, free Android apps were tested on a Samsung Galaxy A54 Smartphone (Samsung Group, Samsung Town,
Seoul, South Korea) using pure tones at 100 Hz, 2000 Hz, and 4000 Hz, across four noise levels (25 dB, 40 dB, and 85 dB). Measurements were compared with a calibrated Precision Gold N09AQ environment meter (Maplin Electronics Ltd, Wath-Upon-Dearne, United Kingdom) in a controlled room. Data were analyzed using linear regression to determine R2 values for each app.

Results

The control meter showed the highest accuracy (R2 = 0.99). SPL Meter dB and sound Meter (KTW apps, Petaling Jaya, Selangor, Malaysia) and Sound Meter (Pony AI Inc.,Guangzhou, Guangdong, China) had the best performance (R2 = 0.98). Accuracy declined at higher noise levels, with Sound Meter (ABC Apps) showing the least accuracy (R2 = 0.85). User ratings did not correlate consistently with app performance.

Conclusion

Android apps offer potential as affordable noise measurement tools, with some apps demonstrating high accuracy at lower decibels. However, limitations such as reduced accuracy at higher decibels and lack of A-weighting for regulatory compliance hinder their use in professional settings. Further development and real-world testing are needed.