Improving sound perception for people with cochlear implants and hearing aids

Background

Hearing aids and cochlear implants improve hearing for people with different degrees of hearing loss, but there are some situations in which listeners using these devices are at a significant disadvantage compared to people with normal hearing.

These situations include listening to speech in noise, music, environmental sounds, and to sounds that are either too loud or too soft.  In particular, most cochlear implant recipients have difficulty appreciating aspects of music, including the perception of pitch and music produced by multiple instruments.

The Institute has completed several research projects with the common aim of improving sound perception for people with cochlear implants and/or hearing aids.  These included basic psychophysical studies to understand the neural and psychological mechanisms of sound perception, development of new sound-processing strategies, and training studies.

Our Research

Improving auditory stream segregation

Music perception and intelligibility of speech in a noisy environment are related to the ability to segregate different sounds according to their source.  People with impaired hearing have reduced abilities for auditory stream segregation compared to people with normal hearing.

We studied the cues used by people with normal and impaired hearing in order to provide cochlear implant and hearing-aid users with enhanced signals which will improve their ability to separate sounds that occur simultaneously but come from different sources. This enabled better perception of speech in noise and music.

Improving music sound quality with cochlear implants

Different instruments have very different sounds. For instance, a trumpet and a violin playing the same note are easily distinguished by people with normal hearing. However, cochlear implant users can find it difficult to distinguish between different instruments, or between sounds that are harsh or smooth.

The ability to distinguish between similar sounds from their timbre is important to recognise a speaker’s voice or a musical instrument. Timbre also conveys emotional information through the harmony and the roughness of the sounds. It is therefore crucial in order to appreciate music fully.

We tested how well cochlear implant users can tease apart melodies played with different types of sounds. This information will be used to develop and test new sound-processing algorithms and training techniques to help cochlear implant users enjoy music more.

Innovative sound processing strategies

Real-time implementations and evaluations of bimodal, STAR, and travelling wave strategies are in progress.  Bimodal refers to people using electrical stimulation from a cochlear implant in one ear together with acoustic stimulation from a hearing aid in the other ear.

Spike-based temporal auditory representation (STAR) is a sound processing strategy based on the way sound is processed in the human auditory system.  Travelling-wave strategies include relative timing information for different sounds as they travel from one end of the cochlea to the other.

Unlike most of the current strategies that stimulate each electrode at the same fixed pulse rate, some of the new strategies present fine timing information, and activate electrodes at different rates or with a modulated amplitude pattern.  Preliminary results showed a significant improvement of the perception of speech in a noisy environment. A processing scheme specifically devised for bimodal device users, called SCORE, is being evaluated with the support of Cochlear Ltd. Initial findings include better understanding of speech when it is heard at lower than ideal levels.
Musical training in children with hearing loss

Soon after receiving a cochlear implant, young children have extensive training in speech recognition and language skills. However, musical training is usually left to the parents, and to the child’s desire to learn. Often a lack of immediate results compared to their friends with normal hearing reduces their motivation to play music. However, if the children and the teacher can anticipate their rate of learning they may be able to adjust their expectations and maintain motivation.  To test this hypothesis, a project was conducted with teacher Christine Storey and a group of children attending Yarra Valley Grammar School.

Multipolar focused electrical stimulation
It is thought that the difficulties implant users experience in appreciating music and understanding speech in noise are due to their poor perception of the fine timing information of sounds. However, this fine temporal information is difficult to convey with existing commercial implant devices and sound-processing methods.
 
In this project, an experimental cochlear implant that can deliver electric currents directly from an external stimulator will be used to overcome the technical limitations of commercial devices. With the support of Cochlear Ltd, we will study how introducing fine timing information across cochlear places can assist recipients to perceive musical pitch accurately, and to understand speech better in noise. These experiments may lead to the development of a new system for clinical application that could provide a substantial improvement for implant recipients.
Bimodal perception of music
A large proportion of patients with a cochlear implant still benefit by using acoustic residual hearing in the non-implanted ear. The two inputs, acoustic via a hearing-aid, and electric via the cochlear implant, provide different information about the sound environment. Usually, the two hearing devices work independently and both are designed to convey as much of the sound signal as possible.
 
We examined how to better convey the sound information via both devices together in order to improve perception of melody, harmony, and other aspects of music.
Spiking neural network model of auditory learning
This project aims to improve sound quality of music and speech in noise by modeling learning processes in the brain.  It is hypothesized that present sound processing strategies for cochlear implants lack the complex temporal patterns that are present in the original sounds, and that prototype sound processors that present this timing information have failed because laboratory experiments do not allow the brain sufficient time to learn the complex patterns. 
 
We used the neural model to assess the “learnability” of new sound processing strategies. Then, we put “easy-to-learn” sound processing into “easy-to-use” processors to give human brains the best chance of adapting to the new sounds they hear through the cochlear implant.  Understanding how and what the brain learns will help maintain Australia’s lead in medical bionics technology, not just for bionic ears, but also bionic eyes, and other applications.
Interior Design: Music for the cochlear implant
Musician, composer, and sound artist Robin Fox, in collaboration with five other composers and the Institute, created a series of six new musical works specially designed for music lovers who have impaired hearing.  Under Robin’s direction, the six composers worked closely with bionic ear recipients, scientists and engineers. The composers responded to the strengths and limitations of the device, and produced works tailored specifically for cochlear implant users.
These works were performed at a concert in February 2011, in front of an audience of 700 people including people with normal hearing and people using cochlear implants. On that occasion, cochlear implant recipients were able to discuss their experiences with normal-hearing music-lovers, perhaps for the first time.
Projects_Hearing_Interior-performers.png

Speak Percussion performing at Interior Design

image description
lines

Make a donation