Sound Classification in Hearing Aids by the Harmony Search Algorithm - Music-Inspired Harmony Search Algorithm: Theory and Applications - page 170

Sound Classification in Hearing Aids by the Harmony
Search Algorithm
Enrique Alexandre, Lucas Cuadra, and Roberto Gil-Pita
Department of Signal Theory and Communications, University of Alcalá,
28805-Alcalá de Henares, Spain
{enrique.alexandre,lucas.cuadra,roberto.gil}@uah.es
Abstract. This chapter focuses on the application of the harmony search algorithms to the prob-
lem of selecting more appropriate features for sound classification in digital hearing aids. Imple-
menting sound classification algorithms embedded in hearing aids is a very challenging task.
Hearing aids have to work at very low clock frequency in order to minimize power consumption,
and thus maximize battery life. This necessitates the reduction of computational load while main-
taining a low error probability. Since the feature extraction process is one of the most time-
consuming tasks, selecting a reduced number of appropriate features is essential, thus requiring
low computational cost without degrading the operation. The music-inspired harmony-search
(HS) algorithm allows for effectively searching adequate solutions to this strongly constrained
problem. By starting with an initial set of 74 different sound-describing features, a number of ex-
periments were carried out to test the performance of the proposed method. Results of the har-
mony search algorithm are compared to those reached by other widely used methods.
Keywords: Sound Classification, Feature Selection, Hearing Aids.
1 Introduction
Hearing aids are electronic instruments that aim at compensating the acoustic loss from
which hearing impaired people suffer. Yet only 20% of hearing-impaired people who
could benefit from digital hearing aids actually purchase one, and of these, about 25%
do not wear the devices [1]! This is believed to be due, at least in part, to the unpleas-
ant amplification of background noise encountered in everyday life. The speech signal
is the primary target and concern for most of users. However, there are many other dif-
ferent sound sources such as traffic, school, people shouting in a football stadium, etc.
Therefore, in order to successfully face the variety of acoustic environments repre-
sented in Figure 1, most hearing aid users generally prefer to have a number of ampli-
fication schemes [2, 3].
A particular application that would be considered as very useful for hearing aid us-
ers, especially by the elderly, is that in which the hearing aid itself classifies the
acoustic environment that surrounds the user, and automatically selects the amplifica-
tion 'program' that is best adapted to such surroundings. This is referred to as self-
adaptation. Some digital hearing aids allow the user to manually select the program
adapted to the acoustic environment he/she is in. This approach commonly exceeds
the abilities of most hearing aid users, especially the elderly. In a second feasible line
of attack, an automatic sound classifier can be embedded in the hearing aid so that it
assists it in selecting the proper program.