Publication:

Cross-platform discrepancies in smartphone sound level meter applications: a multicenter clinical and urban acoustic analysis in Indian context

Background The widespread adoption of tele-audiology necessitates accurate environmental noise monitoring via patient smartphones. This study evaluates the acoustic accuracy and clinical usability of 10 sound level meter (SLM) applications across iOS and Android platforms, comparing performance in controlled audiometric environments against unstructured, real-world Indian settings.

Methods This cross-sectional study incorporates a three-phase design. Phase 1 assessed technical accuracy in a sound-treated booth using pure tones and narrowband noise (250-8000 Hz). Phase 2 evaluated real-world performance, contrasting steady-state noise against impulsive urban traffic noise, highly reverberant outpatient departments (OPD), and a neonatal intensive care uznit (NICU). Measurements utilized a reference class 1 SLM alongside an iPhone 15 Pro, a flagship Android (Samsung S24), and a budget-tier Android. Phase 3 assessed app usability using the mobile application rating scale (MARS).

Results Cross-platform applications demonstrated significant platform-dependent variability. Apps that were accurate on iOS frequently overestimated noise on Android hardware (+3.80 to +4.60 dBA), while native Android apps underestimated noise on the same device (-5.40 dBA), highlighting a software-hardware calibration disconnect rather than uniform hardware failure. Budget-tier Androids exhibited the highest error margins (up to±9.5 dBA), particularly during impulsive urban noise and high-frequency tones (>4000 Hz).

Conclusions Smartphone SLM applications exhibit profound cross-platform discrepancies. Solely validating these tools in controlled laboratories is insufficient for clinical application. To ensure reliable occupational and tele-audiology screening, particularly in resource-limited rural networks, clinicians must utilize standardized biological calibration protocols to derive device-specific correction factors.