Scientists Approve AI-Driven Automated Hearing Test for Remote Use

Researchers from the University of Manchester have developed an AI-powered hearing assessment that can be conducted independently on a computer or smartphone, eliminating the need for clinical supervision. This innovative test leverages advancements in text-to-speech (TTS) and automatic speech recognition (ASR) technologies to create a fully automated and self-administered hearing evaluation.

The core of this technology is a modified Digits-in-Noise (DIN) test, designed to accurately assess human speech understanding amidst background noise. Unlike traditional tests that rely on pre-recorded human speech and require a clinician to interpret responses, the AI version generates speech digitally and uses AI to recognize responses automatically. This allows the entire testing process to be completed within approximately 10 minutes.

In a clinical study involving 31 adults with varying hearing abilities, the AI-based test demonstrated results comparable to conventional DIN tests in both reliability and validity. While some variability was observed, especially among individuals with strong accents, the overall performance was consistent, indicating that AI integration does not compromise accuracy.

The test's design aims to improve accessibility by reducing dependence on specialized equipment and trained personnel, making hearing assessments more convenient and accessible within the comfort of one's home. This approach could be particularly impactful given the long delay—around 8.9 years—that many individuals experience before adopting hearing aids, partly due to hesitation or lack of immediate access to testing.

Experts from the university highlight that integrating machine learning into hearing tests can help overcome existing barriers and make the process more user-friendly, especially for those with language barriers or difficulty using traditional interfaces like keyboards or touchscreens. The research team envisions future iterations of the software incorporating more advanced speech recognition systems to better accommodate diverse accents.

Supported by the National Institute for Health and Care Research (NIHR), the study emphasizes the potential for AI to revolutionize hearing healthcare by enabling reliable, accessible, and personalized testing options for individuals worldwide.

The software developed in this study will be made freely available, paving the way for further innovations in AI-driven audiology and hearing health management.

Source: https://medicalxpress.com/news/2025-10-ai-powered-automated-scientists.html