Through our Flexi Grants, we fund small-scale activities to boost the hearing research field. We support researchers to carry out activities that they’d otherwise struggle to find funding for. This might involve providing funding to carry out a pilot study that will increase their chances of securing larger grants from other funders. Or a grant could support the development of a new collaboration, open up a new area of research, or help researchers to gain valuable new skills.
We awarded our latest round of Flexi Grants last month – and here are some of the new projects we’re supporting:
Improving hearing aid fitting using automatic speech recognition software
The most common treatment for hearing loss is to provide hearing aids, which are programmed to amplify the specific frequencies (tones) that a person may struggle to hear, restoring audibility of the sounds they can no longer hear. However, satisfaction with and use of hearing aids depends on the quality and intelligibility of the amplified sound signals (including speech) provided by the hearing aid. To improve fitting of hearing aids, hearing tests sometimes include measures of how well a person can understand speech through the hearing aids using different programmes. However, these tests are repetitive and time-consuming, causing fatigue and inattention in the people being tested, leading to less accurate results.
To overcome these issues, researchers at the Institute of Hearing Research in Nottingham, and at the Toulouse Institute of Computer Science, are developing automatic speech recognition (ASR) software to test hearing aid settings quickly and reliably. Previous research suggests that ASR software can predict how speech intelligibility declines as hearing loss increases in people, and can interpret speech nearly as well as human listeners, at a range of hearing loss severities. This suggests that in the future, such a system could predict speech intelligibility when listening through a hearing aid, removing the need for time-consuming testing in people, improving hearing aid performance and user satisfaction.
In this project, the researchers will continue to develop and improve the ASR system, and investigate how well it can predict the ability of hearing aids to improve understanding of speech against a noisy background (such as a busy restaurant). Improving fitting of hearing aids in this way would be of huge benefit to people with hearing loss.
Finding an objective measure of tinnitus
Tinnitus, the medical term for any noise, such as ringing or hissing, that people hear in one ear, both ears or in the head but which has no external source, affects over 6 million people in the UK to some degree. It can be debilitating, and is associated with disturbed sleep and depression. There are currently no treatments available to cure tinnitus, just interventions to help people cope.
As yet, we have no way of measuring someone’s tinnitus in an objective way (that is, in a way that isn’t affected by a personal judgment, either by the person with tinnitus or the person measuring it). This makes it very hard to ascertain if what we see in animal studies is relevant to tinnitus in people, as we have no measure that can be used in both animals and people. In addition, any trials of potential treatments in people currently use questionnaires and subjective measures to assess their effects. Such measures are strongly influenced by placebo effects, making them difficult to interpret.
A reliable objective measure of tinnitus severity would greatly enhance the study of tinnitus in people and speed up and improve clinical trials of potential tinnitus treatments. In this project, researchers from the Bionics Institute in Australia will investigate the usefulness of a technique called functional near infrared spectroscopy (fNIRS) as an objective measure of tinnitus, in the hope that they will be able to develop a clinically useful tool to use in tinnitus studies.
fNIRS has many advantages which make it ideal for regular clinical use – it is non-invasive, cheap, and can be used in people with implanted devices like a cochlear implant (unlike some other imaging techniques) – it can also be used with children and infants.
In this project, the researchers will use fNIRS to compare brain activity in groups of people with and without tinnitus, to try to identify brain activity that is specifically linked to tinnitus. If they succeed, they will be able to apply for further funding to develop this measure into a clinical tool that can measure tinnitus objectively. This would speed up and greatly improve how tinnitus clinical trials are carried out and, hopefully, will lead to new treatments sooner.