Image: 

Bilateral signal processing for use in hearing aids and/or cochlear implants

Program: 
R2
Project area: 
R2.2: Bilateral sound processing
Project-ID: 
R2.2.1
Project Status: 
Ongoing

Background

Understanding speech in a noisy background can be challenging, especially for the hearing impaired, hearing aid and cochlear implant users. In normal hearing individuals, the brain is able to 'tune in' to sounds from the direction of the speaker while 'tuning out' to the noisy background sounds - this is not possible if the directionality of the sound is lost, for example when sound is only being processed by one ear.

For hearing aid and cochlear implant users, actively cancelling some of the background noise and providing some directional cues to the sound can address these problems and improve the understanding of speech in such environments.

Detail

This project aims to investigate and develop active noise cancelling technology to improve the understanding of speech in noisy backgrounds for hearing aid and cochlear implant users.

There are two streams to our approach:

1. to investigate and improve bilateral beamformer technology

2. to investigate and develop a strategy that combines bilateral beamformer processing with spatial reconstruction techniques.

One approach to cancellation of noise is to use bilateral beamformer technology. This effectively takes two microphone output signals, one from each side of the head, and combines the information to produce a single 'super-directional' output for the user. This approach to recording and combining sound does not necessarily loose information about the localisation or source of that sound, however the algorithms required to retain this spatial information are too computationally complex for current hearing aid technology.

Combining bilateral beamformer technology with techniques that enable the source or directionality of sounds to be reconstructed will enable this technology to be applied to a new generation of hearing aids and cochlear implants. The 'output' will maintain an appropriate signal/noise ratio for the listener and this can be applied to hearing aids and implants to deliver filtered sound from the environment that still provides the user with directional cues that are important for hearing in noisy backgrounds.