A system for automatic and objective hearing diagnostics in special populations 

In medicine, it is often essential to know to what extent a person can understand speech. Current methods to measure speech intelligibility involve playing specially crafted sentences to a patient, who then repeats what they heard. The results are scored manually by a trained audiologist. The current method requires active patient participation, specialized speech material, and is labor intensive. We present a system to measure speech understanding fully automatically. The system records brain responses (EEG) while the person listens to a recording of natural speech, and applies signal processing to derive speech intelligibility. The measurement is objective since the patient does not need to actively participate. This enables reliable testing of a large variety of clinical populations that currently cannot be tested with traditional methods, such as young children and people with mental impairment. Because our system is low-cost and automatic, it can be applied for diagnostics outside of specialized clinical centers.In this project we aim to bring this system to the market by means of a new spin-off. We will develop a portable product prototype, identify a suitable beachhead market and further validate the system for this target population.

Keywords:Speech intelligibility, Electroencephalography (EEG), Signal processing, Hearing

Disciplines:Audiology