Information Technology Reference
In-Depth Information
population of most highly developed countries is quickly aging, and that hearing im-
pairment has an important impact on communication, society faces a serious health
and social problem. This compels scientists and engineers to develop more advanced
digital hearing aids, and, most importantly, to make them comfortable and easy to
handle, especially for elderly people.
Despite the fact that modern digital hearing aids exhibit the potential to be de-
signed and programmed to compensate for most specific hearing losses, a barrier to
achieving this aim is the user's comfort and subjective preferences. Only 20% of
hearing impaired people who could benefit from a hearing aid acquires the device.
And, as mentioned before, approximately 25% of them do not wear their hearing aids.
What is the explanation underlying these facts? Apart from the high costs of the
most advanced hearing aids (which may provide higher fidelity of sound, greater
overall amplification, directional sound detection, dynamic compression and fre-
quency-specific amplification), there is still a critical, scientific and technological
drawback that has not yet been solved effectively and comfortably at medium-low
cost: the automatic adaptation to the changing acoustic environment.
As shown in Figure 1, this problem arises from the fact that hearing aid users face
a variety of different acoustic environments in their daily life, e.g. quiet conversation,
speech in a noisy cafe, traffic, loud music, etc.
Unfortunately, most of medium-low cost hearing aids are usually designed for only
one listening environment. The overall performance to the individual is unsatisfac-
tory. The unit is either too loud or too soft or the quality is poor. For example, enter-
ing a crowded café produces a sudden, uncomfortable, irritating, and amplified noise.
Many times the patient can 'hear' but not understand the speech signal. With respect
to the diversity of sound environments the patient has to face on in his/her normal life,
it has been shown that hearing aid users normally prefer to have a number of amplifi-
cation schemes fitted to different listening conditions [1-3]. To fulfill such a need
there are two essential approaches in digital hearing aids. The first approach allows
the customer to manually select -among a variety of programs (different frequency
responses or other processing options such as compression methods, directional mi-
crophone, feedback canceller, etc.)- the one the customer considers more adequate to
the acoustic surroundings. The user has to be familiar with the acoustic environment
and determine which program best fits the circumstance by using a switch on the
hearing instrument or some kind of remote control. Nonetheless this approach ex-
ceeds the technical capabilities of most hearing aid users, in particular for the smallest
'in-the-canal' (ITC) or 'completely-in-the-canal' (CIC) hearing aids, which are virtu-
ally hidden to the casual viewer.
The second approach, which could appreciably get better usability of the hearing
aid, is that in which the hearing device itself selects the most suitable program. In this
paper, an HS algorithm is used to select the sound describing features
2 Framework: Design Constraints
As mentioned in the introduction, digital hearing aids present substantial design con-
straints. Figure 2 displays the typical and critical elements of the structure of a digital
hearing aid. The basic components are:
Search WWH ::




Custom Search