Poster: A Machine Learning-Based Approach to Enhance the Accuracy of Sound Measurements in iOS Devices for Accessibility Applications

As technology advances, electronic devices can be a vital part of enabling people with disabilities to live an independent and fulfilling life by facilitating self-sufficiency and social inclusion. Many studies have examined usability challenges of the electronic device context and proposed definitions of application usability and evaluation methods. This study proposes a system solution based on machine learning to minimize iOS device hardware dependencies. The proposed solution analyzes the hardware (microphone) of twenty-five different models of iPhone and twenty-five other models of Pad. The study maps a relation on the hardware changes (microphone) using piecewise linear least-squares fit algorithms. The output of the model is the sound measurement data of all iOS devices acting consistently as sound test devices for the same sound source. Next, integrate this solution with the accessibility measure application named AccessSound (sound level measuring app) and revised the existing algorithm. The proposed solution can contribute to making iPhone and iPad devices more accessible for people with diverse needs, as well as a reference for developers and designers.