Enabling Remote Screening of Childhood Hearing Loss Using Artificial Intelligence

Hearing loss in early childhood can have lasting effects for speech, language, learning, and social development. While universal newborn hearing screening is routine in Australia, many children develop hearing difficulties later in the preschool years that remain undetected. Common conditions such as otitis media with effusion (“glue ear”) can significantly disrupt development during this critical stage.

The Enabling Remote Screening of Childhood Hearing Loss Using Artificial Intelligence project (also known as The Chatterbugs Study), funded by an NHMRC Ideas Grant and conducted by the National Acoustic Laboratories (NAL), addresses this important challenge.

The study is developing an accurate, cost-effective, and accessible hearing screening method for use in the home environment. An in-house designed mobile app enables parents and caregivers to collect speech samples from their children through some fun and easy activities. These samples contribute to a large, nationally representative database including children with confirmed hearing loss, children with glue ear, and children with no hearing concerns.

Using this database, the research team is training advanced deep-learning models to identify subtle speech characteristics associated with hearing loss. This work represents the first study of its kind worldwide, with the potential to establish a new standard for early hearing screening.

By providing a practical tool that can be deployed at scale, the project aims to improve the timely detection of childhood hearing loss, ultimately supporting better outcomes for children and families in Australia and beyond.

This is an ongoing project. This page will be updated at its conclusion. If you are interested in learning more about this project, please reach out to the lead researcher.