Overview

The Hearing Research Lab aims to understand the complex process of listening in noisy environments and then to use this understanding to improve speech perception by people with hearing loss.  As listeners, we have to pull apart the sound we are interested in ("the target sound") from the competing sounds in the environment.  We process sounds in our auditory system and then use higher-level brain proccesses to decode the spoken message.   For people with hearing loss, the process becomes more complicated with the effects of hearing aid technology and the effects of hearing loss added to the mix.   And older adults may also have age-related changes in cognition (e.g., changes in working memory) that may affect the listening process.  The research projects in our lab each addresses one or more of these pieces of the listening chain.

Overivew

Improving Hearing Aid Outcomes

Two adults with similar hearing aids may try the same type of hearing aid but may have very different outcomes.  For one person, it is just the right thing.  For another person, a different type of hearing aid technology may be better.    A major initiative in our lab is to study the sources underlying this variability.    A better understanding as to why listeners with hearing loss vary in their response to different types of hearing aid signal processing will ultimately lead to better ways of customizing hearing aids that are prescribed for people in clinical settings. In collaboration with Dr. Pamela Souza and the Hearing Aid Laboratory at Northwestern University, we are carrying out a set of studies that characterize the listener factors that determine which hearing aid signal processing algorithms lead to the best speech intelligibility and sound quality.   We are considering features such as wide dynamic range compression, frequency lowering, and noise reduction. We are learning that the amount of hearing loss and a person’s age are important. Another major factor is a person’s working memory.  

This work is setting the stage for developing patient specific tests in clinical settings and to provide evidence-based protocols for customizing hearing aid fittings based on individual listener abilities.  This work is funded by the National Institutes of Health.

Perceptual Response to Hearing Aid Signal Processing

An emphasis in our work relates to the design and evaluation of hearing aid signal processing algorithms. The goal of hearing aid technology is to improve the perception of sounds of interest to the listener. The aim to make sounds more audible (can you hear it?), to make sounds more understandable (is it intelligible?), and to make speech or music more pleasant (is the quality good?). In our lab, we evaluate algorithms to see if they are meeting these goals.  Using computer simulations, we measure speech intelligibility and the quality of speech and music processed with the signal processing algorithms.

We also develop metrics that measure how the algorithms impact speech intelligibility, speech quality, and music quality. We send the speech or music signals through a mathematical model of the ear, and we match the model outputs to the data from our listening experiments. The result is a set of metrics that predict how a listener with hearing loss will respond to existing and to new types of signal processing. These metrics are used by our lab and by other researchers to evaluate algorithms as they are being developed.  We are also refining them so that they can be effective tools in the clinical fitting of hearing aids.

Hearing Aids and Remote Microphones

Listeners with hearing loss can have extreme difficulty in understanding speech in noisy or reverberant conditions. One way to improve speech understanding in these situations is to have the conversational partner talk into a microphone and to transmit the remote microphone signal directly to the hearing aid, thereby bypassing the noise and reverberation in the room. There’s a problem, however, because that direct signal also bypasses the acoustic transformations provided by our head and outer ears that we use to locate a source of the sound in space. The result is that the signal appears to come from the middle of our head instead of in front of us in the room. The goal of this research project is to develop computer programs that modify the remote microphone signal so that the apparent location of the remote microphone signal moves back out into the room, thus providing a more natural and comfortable listening experience.  This work is funded by a grant from GN ReSound.

Listening in Multi-talker Environments

Listeners use a variety of cues to navigate auditory environments.  For example, competing talkers are easier to tell apart when they come from different places or when they have different voice pitches.  One line of work in our laboratory focues on understanding the effects of hearing loss and aging -- singly and in combination -- on the ability of listeners to use pitch cues in complex listening situations.   We have explored how older listeners with hearing loss are able to perceive competing talkers who had similar or disparate voice pitches. We have also studied the effects of age and hearing loss on listeners’ ability to use temporal fine structure and periodicity cues in processing fundamental frequency in complex sound environments. Taken together, our research supports the conclusion that age-related declines in multi-talker environments are due to an interaction of degraded peripheral processing and depleted cognitive resources.