Clarity Challenge
Research type
Research Study
Full title
Machine Learning Challenges for Revolutionising Hearing Devices
IRAS ID
276060
Contact name
Michael Akeroyd
Contact email
Sponsor organisation
University of Nottingham
Duration of Study in the UK
5 years, 0 months, 0 days
Research summary
A common first symptom that people notice of hearing impairment is difficulty in understanding speech in noise. This can make communicating in common everyday situations, such as being around traffic, problematic for those with a hearing impairment. Hearing-aids are the typical intervention for this with most forms of hearing loss, but the way in which they process sounds and reduce noise can distort speech making it less clear to understand. Our study will test new hearing-aid technology and software whose purpose is to improve the perception of speech in background noise. This will be achieved by running a series of three “challenges” across a 5-year period, in which research groups around the world will develop new systems to process speech. These challenges are organized by a consortium of researchers at Cardiff, Salford, Sheffield and Nottingham Universities.
Our task at the University of Nottingham is to recruit healthy hearing and hearing impaired participants to mark these sentences for their intelligibility (“what are the words?”). This data will be used to score the different systems and therefore help improve hearing-aid technology.
Participants will be required to spend 20-25 hours across the 5 years completing listening tasks. There are two visits to one of the University of Nottingham’s research centres. During visit one participants complete audiometry testing, various listening and speech identification tasks and questionnaires on hearing in real life scenarios and hearing aid usage (if applicable). During visit two participants receive a demonstration of the provided tablet and “Listen@Home” software, which loads the listening tasks and records the responses automatically, so that the remaining tasks can be completed at home by the participants. During this visit participants will complete a “dummy run” of the intelligibility marking task to ensure they’re comfortable with the device and Listen@Home.
REC name
London - London Bridge Research Ethics Committee
REC reference
20/LO/0666
Date of REC Opinion
4 May 2020
REC opinion
Favourable Opinion