“Self learning” or trainable hearing aid

Researcher(s).: M. Boymans, PhD.
Supervisor(s): prof. W.A. Dreschler, PhD., in cooperation with

H. Dillon, PhD., and

G. Keidser, PhD. (NAL, Sydney).

Background:
Advanced hearing aids are able to react interactively to the dynamical environment of the hearing aid user. The possibilities to profit fully from this interaction is hampered by lack of knowledge about the way dynamical hearing aids should be fitted to the properties of the ear of the individual user and his or her specific wishes and preferences. This can be understood from the fact that there are different types of hearing losses, there is a large variety of relevant sound characteristics and the acoustical environment has a large impact. Last but not least the individual preference about the soundscape varies among subjects.

Aim:
The goal of this project is to investigate if a trainable hearing aid is able to give full control to the real expert: the hearing aid user. This project is initiated by the National Acoustical Laboratories in Sydney, Australia.

Method:
In a laboratory set up (conducted at NAL) we investigated if hearing aid users are able to adapt the frequency gain characteristics reliably to different acoustical situations. Experimental questions are which type of controller is the best, what is the reliability of the user settings, and what is the opinion of the hearing aid user about the possibility to train the hearing aid him of herself.
In a field trial we investigate if settings for gain and compression can be trained in the prototype trainable hearing aid.

Results:
The laboratory study was conducted with 24 hearing aid users (NAL in Sydney). The results show that:

The most important parameters for gain as a function of frequency can be selected successfully by the hearing aid users.
For different acoustical situations significant differences are found.
The reliability is in the order of 2.5 to 3 dB, depending on the controller used.
The majority of the users likes the idea to have their own control about the hearing aid and should like to have a controller available during a trial period with new hearing aids.

Future Research:
The results of the field trial will be important for a successful application of this concept in daily practice.


Home Onderwijs Wetenschap Research (EN)
General Otology Prevention of hearing loss Auditory profile Rehabilitation Evaluation of Noise Reduction in (digital) Hearing Aids AZOS-project: Modified care model for hearing impaired listeners Classification of hearing aids in CLiQ Interactive fitting of hearing aids “Self learning” or trainable hearing aid Evaluation of advanced hearing aid features Optimalisation of hearing aids through paired comparisons Field trials with innovative hearing instruments The added value of a bilateral hearing aid fitting Hearing at work Publications Presentations	“Self learning” or trainable hearing aid Researcher(s).: M. Boymans, PhD. Supervisor(s): prof. W.A. Dreschler, PhD., in cooperation with H. Dillon, PhD., and G. Keidser, PhD. (NAL, Sydney). Background: Advanced hearing aids are able to react interactively to the dynamical environment of the hearing aid user. The possibilities to profit fully from this interaction is hampered by lack of knowledge about the way dynamical hearing aids should be fitted to the properties of the ear of the individual user and his or her specific wishes and preferences. This can be understood from the fact that there are different types of hearing losses, there is a large variety of relevant sound characteristics and the acoustical environment has a large impact. Last but not least the individual preference about the soundscape varies among subjects. Aim: The goal of this project is to investigate if a trainable hearing aid is able to give full control to the real expert: the hearing aid user. This project is initiated by the National Acoustical Laboratories in Sydney, Australia. Method: In a laboratory set up (conducted at NAL) we investigated if hearing aid users are able to adapt the frequency gain characteristics reliably to different acoustical situations. Experimental questions are which type of controller is the best, what is the reliability of the user settings, and what is the opinion of the hearing aid user about the possibility to train the hearing aid him of herself. In a field trial we investigate if settings for gain and compression can be trained in the prototype trainable hearing aid. Results: The laboratory study was conducted with 24 hearing aid users (NAL in Sydney). The results show that: The most important parameters for gain as a function of frequency can be selected successfully by the hearing aid users. For different acoustical situations significant differences are found. The reliability is in the order of 2.5 to 3 dB, depending on the controller used. The majority of the users likes the idea to have their own control about the hearing aid and should like to have a controller available during a trial period with new hearing aids. Future Research: The results of the field trial will be important for a successful application of this concept in daily practice.