Listening to Music with Headphones: An Assessment of Noise Exposure

Listening to Music with Headphones: An Assessment of Noise Exposure
Listening to Music with Headphones:
An Assessment of Noise Exposure
and Hearing Damage
Department of Acoustics
Aalborg University
Beatriz Gutiérrez Camarero
Irene Moledero Domínguez
Institute of Electronic Systems
Department of Acoustics
Aalborg University
Title:
Abstract:
Headphone Sound Exposure and Hearing
Many studies suggest that music players used with
headphones can emitt sound pressure levels as high
as the noises emitted in some industries and working
places. Both exposures might be harmful to hearing
in prolongate use. Moreover the degree of damage
produced by listening to music with headphones is
not clearly defined. This has been the motivation
for this project which is to investigate the posible
presence of a permanent hearing damage due to
headphone sound exposure.
Project period:
ACO-8
5. Febrary - 7. June 2007
Projectgroup:
1066
Group members:
Beatriz Gutiérrez
Irene Moledero
In order to fullfil this aim, two populations (control
and target) which differ significantly in their habits
when listening to music from headphones are compared using a headphone sound exposure evaluation
an a hearing assessment. This is carried out by testing
20 selected subjects in a listening test.
Supervisor:
Rodrigo Ordoñez
This test consists of a pure tone audiometry, a
DPOAE measurement and an analisis of the exposure
level of the subjects in different environments.
It is concluded that the results obtained could not describe a potential damage by listening to music from
a MP with headphones. Moreover, the environment
around a person affects the preferred volume setting
that this person select in his MP specially when the
environment becomes noisy. However, users of MPs
with headphones tent to listen to music at higher volume settings than the non users.
Number of copies: 4
Report – number of pages: 59
Appendix – number of pages: 87
Total number of pages: 146
3
P REFACE
This report was written by group 1066 of the 10th semester as the documentation of the Thesis of the international Master of Science programme in Acoustics at the Institute of Electronic
Systems at Aalborg University.
The report is primarily addressed to students and staff of the Department of Acoustics at Aalborg University, and anyone interested in the possible noise inductive hearing loss due to the
exposure to music from music players with headphones.
The report is divided in two parts; main report and appendix. The main report is divided into
the next several chapters:
• Chapter 1 gives an introduction to the reader and sets the framework for project focus.
• Chapter 2 presents an overview of the background theory related to the topic of the
project.
• Chapter 3 introduces an analysis describing the most important considerations of the
project.
• Chapter 4 exposes a pilot test carried out to define some parameters used for a listening
test.
• Chapter 5 exposes the listening test carried out to investigate the topic of the project.
• Chapter 6 presents the results obtained from the listening test and their analysis.
• Chapter 7 summaries the work done in the project and the results obtained.
The appendix part includes descriptions of measurements, supplementary documentation and
important issues regarding the project.
A CD-ROM is enclosed. It contains the report in PDF-file format, the data obtained during the
experiment, the Matlab code, the sample files, articles and data sheets.
We would like to thank Miguel Angel Aranda de Toro and Juan Luis Mateo, for all the help
given to us during the project; Claus Vestergaard Skipper for their technical assistance and all
the subjects that participated in the experiment for their patience and good will.
5
Aalborg University, June 7, 2007
Irene Moledero
Beatriz Guiterrez
6
C ONTENTS
I
Report
1
1
Introduction
3
1.1
The Aim of the Project . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
4
1.2
Scope of the Project . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
4
2
3
4
5
Hearing and Sound Exposure
5
2.1
Hearing and Sound . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
5
2.2
Hearing System anatomy . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
6
2.3
Noise Inductive Hearing Loss (NIHL) . . . . . . . . . . . . . . . . . . . . . .
9
2.4
Hearing Assessment . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
11
2.5
Audiometry . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
11
2.6
OtoAcustic Emissions (OAE) . . . . . . . . . . . . . . . . . . . . . . . . . . .
13
2.7
Selection of the Assessment Techniques for NIHL . . . . . . . . . . . . . . . .
16
2.8
Chapter Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
19
Headphone Sound Exposure
21
3.1
Listening Devices: Music Players and Headphones . . . . . . . . . . . . . . .
21
3.2
Listening Source: Music . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
22
3.3
Headphone Sound Exposure Evaluation . . . . . . . . . . . . . . . . . . . . .
22
3.4
Headphone Sound Exposure Parameters . . . . . . . . . . . . . . . . . . . . .
24
3.5
Chapter Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
25
Pilot Test
27
4.1
Preliminary Survey . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
27
4.2
Pilot Test Design . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
27
4.3
Chapter Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
32
Listening Test
35
i
CONTENTS
6
7
5.1
Goals of the Listening Test . . . . . . . . . . . . . . . . . . . . . . . . . . . .
35
5.2
Listening Test Design . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
35
5.3
Listening Test Subjects . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
36
5.4
First Part: Listening Test Questionnaire . . . . . . . . . . . . . . . . . . . . .
36
5.5
Second Part: Hearing Assessment . . . . . . . . . . . . . . . . . . . . . . . .
36
5.6
Third Part: MP Volume Adjustment . . . . . . . . . . . . . . . . . . . . . . .
37
5.7
Chapter Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
38
Listening Test Results
39
6.1
Listening Test Analysis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
39
6.2
Analysis of Listening Level . . . . . . . . . . . . . . . . . . . . . . . . . . . .
40
6.3
Analysis of Exposure Level . . . . . . . . . . . . . . . . . . . . . . . . . . . .
43
6.4
Analysis of Hearing Thresholds and DPOAEs . . . . . . . . . . . . . . . . . .
45
6.5
Chapter Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
52
Conclusions
55
Bibliography
57
II
61
Appendix
A Assessment Techniques Parameters
63
A.1 Audiometry . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
63
A.2 DPOAE Measurements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
63
B Types of headphones
65
B.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
65
B.2 Headphone Categories . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
65
C Exposure Level Calculation
67
C.1 Method . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
67
C.2 Setup and Equipment . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
68
C.3 Results . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
68
D Pilot Test Measurements
73
D.1 Method . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
ii
73
CONTENTS
D.2 Setup and Equipment . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
73
D.3 Results . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
75
E Pilot Test Instructions
83
F Pilot Test Questionnaire
85
G Pilot Test Subjects
93
H Listening Test Instructions
97
I
J
Instructions for the Test Experimenter
101
I.1
Basic Rules for the Test Experimenter . . . . . . . . . . . . . . . . . . . . . . 101
I.2
Test Experimenter Instructions . . . . . . . . . . . . . . . . . . . . . . . . . . 101
Listening Test Questionnaire
105
K Listening Test Subjects
107
L Hearing Threshold and DPOAEs
115
L.1 Pure Tone Audiometry . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 115
L.2 DPOAEs Measurements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 115
L.3 Results . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 116
M Listening Environments
131
M.1 Method . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 131
M.2 Equipment and Setup . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 131
N ANalisy Of VAriance (ANOVA)
137
O Independent Samples T-Test Analysis
139
P Statistical Analysis
145
iii
Part I
Report
1
C HAPTER 1
I NTRODUCTION
Over time, many scientific studies have investigated the potentially harmful effects of noise and
its consequences. Many of them are focus on the hearing damage caused on adults that are
exposed to a noisy environment at their working places [4][29][8]. Nevertheless, there are also
leisure time activities that can produce hazardous noise levels as well [9][30][39]. These leisure
exposures are for example: the sound emitted by some electronic devices such as toys or mobile
phones, the sound in cinemas, the music in concerts etc.
Music players (MPs) used with headphones is other of these leisure activities. The use of these
devices offer a convenient way to listen to music as high listening level as the user likes without
disturbing others. Some of these MPs are portables, for instance MP3 players or pocket computers. Nevertheless there are many others which are not, such as televisions or Hi-Fi equipments.
One of the main reasons of the increasing of users of MPs with headphones is the popularity of
MP3 players in the last years. The reason of this success is its portability. Moreover, as technology improves, with greater music storage and longer battery life, it is possible that people will
choose to listen for longer periods of time than ever before.
Scientific studies suggest that these MPs, portable or not, may cause hearing damage if they are
not used with a degree of caution [10][41]. The exposure time (how often, how long) and the
listening level are the main factors that may influence the possible hearing damage [42]. Moreover, it seems that the acoustic environment around a person may affect the listening level that
a person sets in his MP [10][11].
Investigations about the possible hazardous effects of headphone sound exposure have been
made before, however they have yielded to different results [9]. Some researchers conclude that
personal stereo systems pose a risk [9] whereas others claim that their effect is not so harmful
[44][3]. In general, it is agreed that there is some level of possible risk to hearing in certain
conditions. This leads to the aim of the project.
3
CHAPTER 1. INTRODUCTION
1.1 The Aim of the Projet
The aim of the project is to investigate the possible hearing damage due to the use of MPs
with headphones. In order to do that, two groups of subjects which differ significantly in their
listening habits are analyzed by means of an evaluation of headphone sound exposure and a
hearing assessment.
1.2 Sope of the Projet
The scope of the project is to:
• Describe the needed theory in order to fulfill the aim of the project. This includes theory
regarding the human hearing, hearing disorders and hearing assessment.
• Identify two groups of subjects denoted as control and target population. The target population is characterized by an exposure to music from MPs with headphones at high listening levels and during long periods of time. By contrast, the control population is defined
as a group which do not use, or use MPs with headphones at not risky levels during very
short periods of time.
• Design and carry out a listening test that consists of a hearing assessment and a headphone
sound exposure evaluation of the populations defined before. Different environments to
which these populations are exposed to when using MPs with headphones are considered.
• Carry out the acoustic measurements of different MPs used with headphones in terms of
their ability to produce sound pressure levels.
4
C HAPTER 2
H EARING AND S OUND E XPOSURE
This chapter reviews basic information about sound, human hearing, and how sound exposure
affects to the human hearing. Moreover some techniques to assess the ability to hear sounds are
explained. The chapter is concluded with the decisions made for the project based on the theory
explained.
2.1 Hearing and Sound
Sound can be defined as pressure variations in the air created by vibrating objects and propagated through a medium from one location to another. These pressure vibrations are called
sound pressure waves. The hearing process occurs when sound pressure waves reach the ears
of a listener. The auditory system is the responsible of this hearing process, which is based on
a conversion of the sound pressure waves to impulse signals. Then these impulsive signals are
transmitted to the brain where the sound is perceived.
In terms of amplitude, the hearing of a person is limited by the next two definitions: the threshold of hearing and the threshold of pain. The smallest intensity of a sound that a person needs
to detect its presence is called hearing threshold [13], whereas the threshold of pain refers to the
intensity of a sound stimulus at which it becomes interpreted as painful.
In terms of frequency range, a person is able to hear frequencies from 16 to 20 kHz [7]. However, a person is not equally sensitive at all frequencies, thus the threshold of hearing varies from
one frequency to another. Different hearing thresholds are expected if the acoustics environment
or the technique used to measure the threshold are modified.
Furthermore, people is not equally sensitive, which means that the threshold of hearing varies
among persons. However, standard thresholds values can be derived from a group of otologically normal persons. This concept is defined in the standard ISO 8253-1(1989)[26] as: a person
who is free from all signs or symptoms of ear desiase and from obstructing wax in the ear canal,
and who has no history of undued exposure to noise.
Figure 2.1 shows the standard thresholds values presented in ISO 389-7 [24]. It is observed
that the frequencies where the ear is more sensitive is at middle frequencies, between 500 and
5000 Hz [37].
5
CHAPTER 2. HEARING AND SOUND EXPOSURE
100
Free field
Diffuse Field
80
SPL [dB ref. 20 µ Pa]
60
40
20
0
−20
20
100
1000
Frecuency [Hz]
10000
Figure 2.1: Minimum audible sound pressure level (thresholds values) presented in the standard ISO 389-7
[24℄ in free eld and diuse eld of an otologially normal person in the age range from 18 years
to 25 years. The data is plotted as a funtion of frequeny. The proedure used for obtaining
this data is dened in the standard ISO 8253-1 [26℄.
It can be observed in Figure 2.1 that the minimum sound pressure level that a person can detect
at 1000 Hz corresponds to 20 µPa. All the sound pressure levels expressed in decibels (dB) in
this report are referenced to this value.
The sounds that a person is exposed to, may modify his hearing threshold under certain circumstances. This sound exposure level is denoted by LEX,To according to ISO 1999 [23], which is
applied for determination of occupational noise exposure. LEX,To combines the listening level
and the exposure time to which a person is exposed during a reference time of To hours.
In order to analyze how sound exposure can affect the hearing, some aspects related to the
anatomy and physiology of the auditory system are explained in the next section.
2.2 Hearing System anatomy
The auditory system is divided into three main parts: the outer ear, the middle ear and the inner
ear, as it is shown in Figure 2.2. The outer ear is composed of the pinna, the ear canal and
the tympanic membrane. The pinna is the external cartilaginous part with asymmetrical and
irregular shape. The middle ear is an air-filled cavity that is composed of three small bones,
the ossicles. This part of the auditory system is connected via the Eustachian tube to the nasal
part of the pharynx. Finally, the inner ear comprise the cochlea and the auditory nerve through
which the impulse signals are sent to the brain.
6
2.2. HEARING SYSTEM ANATOMY
2.2.1 Hearing proess
The hearing process takes place in the three main parts that it is made up:
Outer ear
The incoming sound waves are filtered by the pinna together with the head and torso. Then the
sound travels along the ear canal to reach the eardrum which starts to vibrate as the membrane of
a microphone would do. These vibrations are transmitted though the middle ear to the cochlea
by the ossicles.
Middle ear
The major function of the middle ear is to ensure the efficient transfer of sound from the air
to the incompressible fluids that are inside the cochlea. If the incoming sounds were applied
directly on the entrance of the cochlea, which is named oval window, most of them would be
simple reflected back instead of enter into the cochlea. It happens because there is a difference in acoustical impedance between the low impedance of the air in the eardrum and the
high impedance of the fluid inside the cochlea. The main function of the ossicles is to act as
an impedance adaptor, coupling the two different acoustical impedances. Thus, this function
together with the difference in area between of the eardrum and the oval window, improve the
sound transmission. Moreover the amount of reflected sound is also reduced.
Figure 2.2: Deomposition of the hearing organ in three main setions: the outer ear (A), the middle ear
(B) and the inner ear (C). Figure adapted from [13℄.
Transmission of sound through the middle ear is most efficient at middle frequencies. In addition, at low frequencies the sound transmission through the middle ear is reduced due to some
7
CHAPTER 2. HEARING AND SOUND EXPOSURE
contraction caused by the minute muscles attached to the ossicles. This contraction is known
as middle ear reflex and it helps to prevent damage to the delicate structures of the cochlea.
However this reflex is too slow to provide any protection against impulsive sounds, such as gun
shots or hammer blows.
Inner ear
Once the sound is transmitted through the middle ear, it reaches the cochlea via the oval window. The cochlea is a bony structure shaped like the spiral shell of a snail. Within of it there
is the incomprehensible fluid which acts as a medium to conduct the mechanical vibrations into
pressure waves.
The colchea is divided into three cavities: the Scala vestibuli, the Scala media and the Scala
tympani. The Scala media is bounded by two membranes: Reissnert’s membrane and Basilar
Membrane (BM).
The start of the cochlea where the oval window is situated is known as base while the other end
is named apex. When the oval window is set into motion, a pressure difference is applied across
the BM. As a consequence of this effect, a wave traveling from the narrow base toward the wide
apex of the BM is formed. The amplitude of the sound wave created in the BM increases at first,
then when the maximum peak is reached, the sound wave decreases abruptly.
The localization of the maximum peak depends on the frequency content of the incoming sound
wave. High frequency sounds produces the maximum displacement of the BM close to the base.
By contrast, low frequency sounds set the whole membrane into motion and reaches the maximum near the apex. The BM is behaving as a frequency analyzer, each place on it, is sensitive
to a narrow frequency range. It can be appreciated in Figure 2.3 the shape of the traveling wave
and the location of its maximum according to the frequency content.
1600 Hz
400 Hz
100 Hz
Basilar membrane
displacement
0
5
10
15
20
25
30
35
40
45
50
55
60
Apex
Base
Distance from stapes along basilar membrane (mm)
Figure 2.3: Displaement along the BM for three dierent frequenies.
8
2.3. NOISE INDUCTIVE HEARING LOSS (NIHL)
2.2.2 The role of the hair ells
Inside the Scala media, all along the BM are hair cells which form part of a structure named
the Organ of Corti. Figure 2.4 shows a cross section of the Organ of Corti, where two types of
auditory cells can be appreciated: the inner hair cells and the outer hair cells.
Figure 2.4: Cross setion of the Organ of Corti whih ontains the inner and outer hair ells. Figure obtained
from [1℄.
The inner hair cells are responsible for the transduction of mechanical movements into neural activity. These cells are connected to afferent neurons which carry informations from the
cochlea to high levels of the auditory system [37].
The outer hair cells are connected to efferent neurons, which carry information from the brain to
the cochlea. The main role of the outer hair cells is to enhance the frequency selectivity and also
to produce a non-linear amplification process at low levels in the BM [37]. However, practically
no reaction is produced at more intense sounds. This non-linear function is easily damage by
noise. Next section explains this type of hearing disorder.
2.3 Noise Indutive Hearing Loss (NIHL)
Noise-Inductive Hearing Loss (NIHL) is a sensorineural hearing disorder referred to as permanent damage, cause by noise, to the outer hair cells resulting in a decreasing of the amplification
ability of the cochlea [8]. Anatomical changes such as the fusion or disappearance of the hair
cells are the main effects that causes this decreasing[8].
The outer hair cells are generally more susceptible to damage than the inner hair cells. Mainly,
the ones corresponding to the frequencies between 3 to 6 kHz are affected [13]. Nevertheless
if a person continues being exposed to harmful noises, the hearing damage spreads over other
frequencies as well.
9
CHAPTER 2. HEARING AND SOUND EXPOSURE
NIHL is caused due to an over stimulation of the auditory system produced by an one-time exposure to loud sound, as well as by repeated exposure to sounds at various loud levels over an
extended period of time. Therefore, the effects of these over stimulations can be seen slowly
over years of continuous exposure, or instantaneously after one-time exposure to loud sound. In
the case of an overall year exposure, the full effects of NIHL are generally noticed after ten or
more years of noise exposure [8].
Furthermore, there are other reasons that can produce a damage in the hearing cells as an inner
ear infection due to external bacteria, an ingestion of ototoxic drugs or inherited conditions.
NIHL is a preventable hearing disorder that affects people of all ages and demographics because
it depends mainly of the listening habits, the noisy environments that a person is exposed to, and
the characteristics of the noise. One of the possible consequences of noise exposure is tinnitus.
Tinnitus is defined as the abnormal perception of sounds for which there is no external stimulus
[13]. It is usually to be perceived in one or both ears or in the head. Some people define it
as a ringing noise, a buzzing or a whistling sound. This phenomena can be suffered also by
otollogically normal persons after a sound exposure.
It has to be noticed the difference between the NIHL and the hearing loss associated with advancing age, which is called presbyacusis. This hearing loss may be due to lesions in the external
or middle ear, but the most consistent effect of aging is on the hair cells and neurons [33].
2.3.1 Hearing Loss Desriptors
There are some descriptors associated with the hearing loss explained in this section. These
descriptors are combined with the NIHL to give additional information. Some of them are:
• Bilateral versus unilateral hearing loss:
Bilateral hearing loss occurs when both ears are affected, by contrast unilateral hearing
loss means that only one ear is affected.
• Symmetrical versus asymmetrical hearing loss:
The difference between symmetrical and asymmetrical hearing loss is based on the degree
of hearing loss in each ear. If both ears are damaged in the same way, the hearing loss is
symmetrical. Otherwise it is asymmetrical.
• Progressive versus sudden hearing loss:
A hearing loss is progressive when the damage produced in the ear increased over time.
However, a sudden hearing loss is a ear damage that occurs in a very short period of time.
• Temporary versus permanent hearing loss:
10
2.4. HEARING ASSESSMENT
A temporary hearing loss or temporary threshold shift is defined as a change in the hearing thresholds that disappears within a period of time after the exposure. By contrast, a
permanent hearing loss is a hearing disorder that does not vary over the time.
These descriptors of the NIHL can be identify using different methods. Next section explains
the methods utilized for hearing assessment in this project.
2.4 Hearing Assessment
The purpose of hearing assessment is to quantify the ability to hear sounds. A variety of methods
can be used, such as audiometry and OtoAcoustic Emissions (OAE).
2.5 Audiometry
This test measures hearing thresholds by means of behavioral feedback from the subject. The
person taking the test is instructed to give some type of response when a sound stimuli presented
is heard. There are different possibilities for the subject to express if the sound is perceived such
as pressing a button or raising a finger. The response given by the subject is caused by the
sensory impression of the sound stimuli, but also the interpretation based on the experience and
knowledge of the test subject influences in the decision of hearing or not a sound [34].
During this test, headphones are usually used. Thus sound travels through the air in the ear canal
to stimulate the eardrum and then the auditory nerve. This procedure, which is specified in the
standard ISO 8253-1 [26], is called air conduction audiometry.
Thresholds are measured for different frequencies in each ear. The responses are recorded on a
graph called audiogram that provides the Hearing Level (HL) for each frequency tested. The HL
expresses the difference in decibels between a measured hearing threshold and the thresholds
values derived from an otologically normal population at a particular frequency [25]. A straight
horizontal line at 0 dB HL in the audiogram represents the hearing derived from these thresholds
values. For example if a person has a hearing threshold of 25 dB HL at a specific frequency, it
means that the hearing threshold is 25 dB higher than the threshold of hearing obtained from a
population considered otologically normal at that specific frequency.
Types of Audiometries
There are many different types of audiometries for threshold determination. For instance the
method of limits, the Békésy method, the method of constant stimuli, forced choice methods,
audioscan method, the method of adjustment etc.
11
CHAPTER 2. HEARING AND SOUND EXPOSURE
Many of them gradually converge on the threshold by presenting sound stimuli at levels that
depend on the response to previous trials, considering a trial as the answer given by the subject
in each stimuli presentation. After several trials of one subject, the hearing threshold is determinated according to a stop criteria.
The methods to be analyzed in detail for this project are the method of Limits and the Bekesy
method. Both methods are standardized in ISO 8253-1 [26].
Method of Limits
There are different versions of this method: Ascending, Descending and Bracketing. They all
are based on the same principle: several pure tones at specific frequencies are presented to the
subject, so by means of a variation in the level of these pure tones the hearing threshold is
determined.
• Descending Method
It consists of a presentation of several pure tones starting from a level which is above
to the threshold of hearing. Then the subject must show if the sound was heard or not.
Since the starting level is above enough to the threshold of hearing, a positive answer is
expected. Then the level of the pure tone is decreased and presented again to the subject.
This process is repeated until the subject is not able to perceive the sound. Every time that
the subject finishes a descending from an audible to an inaudible stimuli is called a run
[14]. Several runs are performed, so the threshold of the subject is obtained according to
the threshold levels across runs [14]. This is made for each frequency.
• Ascending Method
The ascending method differs to the descending method in the starting level of the pure
tones. In this case the sound that is first presented is set to a level quite below to the
threshold of hearing. Then the level of the presented pure tones is increased until the
subject hears the sound. Again several runs are performed and the threshold is calculated.
• Bracketing Method
This method combined the ascending and descending methods. Starting from a level
above the threshold of hearing, the stimuli is decreasing until the subject can not perceived. Then the stimuli is increased again until the subject shows that can hear the
sound. Therefore a combination of increasing and decreasing runs are performed when
using this method. This process is repeated a certain number of times for each frequency.
Then from the values obtained in the descending and ascending runs, the threshold is calculated.
The drawbacks of this method are the anticipation and habituation of the subject to the task.
Sometimes subjects anticipate hearing the stimulus and sometimes anticipate not to hear it.
12
2.6. OTOACUSTIC EMISSIONS (OAE)
Other disadvantage is that is very dependent on the amount of decibels when increasing or decreasing the levels. A big increasing or decreasing in the runs can lead to inaccurate results.
An advantage of this method is that a full range of performance levels can be estimated. However, if only hearing threshold determination is desired, many trials will be presented until reach
the hearing threshold.
Békésy Method
In this method the stimulus is controlled by the test subject who adjusts the level continuously
over the time to reach the hearing threshold. The sound stimuli are pure tones as well as in the
method of Limits, but the task of the subject is different. In this case the pure tone is presented
and the subject has to press a button when the tone is heard and to release it when the signal is
inaudible. When the subject release the button, then the level of the tone increases automatically
in a certain step size. Therefore the level is increased and decreased several times depending on
the answers of the subject. This will produce some sort of valleys and peaks around the minimum audible level. The hearing threshold is calculated according to certain number of peaks
and valleys.
A disadvantage of the Békésy method is that the measurements depends on the reaction time of
the subject. If the stimulus are decreasing or increasing so fast that the subject does not have
time enough to press or depress the button according to his perception of the sounds, then the
thresholds obtained are not reliable. This can be avoided using reasonably slower attenuation
rates.
On the other hand, this method is fast an has the advantages of speed and reasonable precision.
because of the tracking used to achieve the hearing threshold.
2.6 OtoAusti Emissions (OAE)
Besides the hearing process, a healthy human ear is able to produce inaudible sounds emitted by
the cochlea which are known as OAEs. They can be produced either spontaneously or by means
of a stimulation of the ear with one or several specific sound signals.
The OAEs are produced in a non damage cochlea by its own activity and reflect the activity of
the outer hair cells. The sounds emitted by the ear are in fact produced due to a mechanical
motion of the outer hair cells of the cochlea that is transmitted to the basilar membrane [31].
This active mechanical process is consequence of natural imperfections of the cochlea amplifier
mechanism [43]. The motion of the hair cells creates a vibration which is driven by the cochlea,
passing through the oscicles and finally reaches the tympanic membrane. Then a vibration of
13
CHAPTER 2. HEARING AND SOUND EXPOSURE
the tympanic membrane occurs and a sound wave is emitted by the ear. If a probe microphone
is placed into the ear canal closing it, then this sound wave can be recorded.
Although the position of the probe microphone the amplitude of the OAEs may vary slightly
[45], some studies have investigated how the amplitude of the OAE varies depending on the
degree of hearing loss[43][28][8][12] [31][32]. Most of them conclude that for a population the
presence of OAEs is related to normal hearing thresholds, whereas reduced OAE levels can be
associated with hearing loss.
Types of OAEs
The OAEs can be classified as evoked or spontaneous OAEs. Evoked OAEs are inaudible
sounds emitted by the cochlea when the cochlea is stimulated by a sound signal. When this
sound reaches the ear, it is transmitted through the ear canal and the oscicles until the cochlea.
Then the outer hair cells vibrate creating an inaudible sound that echoes back into the middle
ear again, which is the Evoked OAEs. By contrast, if the ear produces the vibration of the hair
cells by itself in the absence of any know stimulus, then the OAEs are called spontaneous OAEs.
Among the evoked OAEs, there are different types depending on the acoustic stimuli used. They
are transient evoked, stimulus frequency and distortion product OAEs.
Transient Evoked OAEs (TEOAEs)
These OAEs are produced when a broadband signal is presented as stimuli, for example a click
or a tone burst. The whole cochlea is activated from the apex to the base, so a broadband
frequency region is tested when measuring TEOAEs. The response of the ear to this stimuli is
long and complex because responses from different parts of the cochlea arrive at the ear canal
at different times and at different frequencies [31].
Stimulus Frequeny OAEs (SFOAEs)
In this case a narrow band stimulus is applied, for example a pure tone. This type of stimuli
activates a narrow zone in the cochlea. To excite a wider zone successive stimulations are
needed.
Distortion Produt Otoaousti Emissions (DPOAEs)
If the sound stimuli used is based on a multi-tone stimuli, for example two sinusoids of different frequencies f 1 and f 2 ( f2 > f 1 ) and levels L1 and L2 , then the cochlea generates several
new acoustic frequency components besides the frequencies f 1 and f 2 . These new frequency
components, called distortion products, are due to a non-linear intermodulation between the
14
2.6. OTOACUSTIC EMISSIONS (OAE)
two stimulus tones along the basilar membrane. In the case of the two sinusoids, the distortion
products are 2 f 1 − f 2 , 3 f 1 − 2 f 2 , 2 f 2 − f 1 etc. They are obtained according to Equation 2.1
f d p = f 1 + N( f 2 − f 1 )
(2.1)
where N is any positive or negative whole number [31].
It is shown that the intermodulation distortion produced at the frequency 2 f 1 − f 2 is the DPOAE
which can be measured easily due to its amplitude, although the cochlea also produces DPOAEs
at other frequencies, as it was mentioned before. In order to have a good response at that frequency, the levels and the frequencies of the tones must be selected adequately. An f 1 / f 2 ratio
at 1.2 and a intensity of 65 and 55 dB sound pressure level for L1 and L2 respectively yields the
greatest DPOAEs [31].
Figure 2.5 shows the development of the traveling waves f 1 and f 2 along the BM. In graph a)
corresponds to a dead cochlea. It can be seen that the cochlea does not respond to most of the
stimulus energy. In graph b) a cochlea with a linear amplification of the outer hair cells is shown.
However the outer hair cells do not have a linear behavior, and this results in intermodulation
distortion products (2 f 1 − f 2 and 2 f 2 − f 1 ) which then travel to their frequency places at points
3 and 4 and generate the backward traveling waves. This can be seen in graph c). Therefore the
DPOAEs generation is based on a two-source model: The first source is the initial non-linear
interaction of the primaries and the second source comes from the re-emission site at the characteristic place of the distortion product frequency. If a decreased DPOAE is measured, it may not
necessarily mean that the damage area of the BM corresponds to the place where the 2 f 1 − f 2
and 2 f 2 − f 1 are generated. It could be by a BM damage in the overlap area between f 1 and f 2
for example.
By modifying the frequency of f 1 and f 2 , the distortion product 2 f 1 − f 2 varies, therefore different frequencies can be tested. If the ratio between f 1 and f 2 or the intensity of the tones changes,
the DPOAEs can not be compared among them in terms of amplitude. For this reason, these
two parameters have to be maintained constant all along the frequency range.
Depending on the frequency variation when changing f 1 and f 2 along the frequency range,
different resolutions can be obtained. For example, if a frequency variation of 15 Hz is used,
many values are obtained for a specific range of frequencies. In this range peaks and valleys
are detected when measuring the DPOAEs because of the high resolution. Nevertheless, if the
frequency variation is selected for giving DPOAEs values each 1000 Hz, then these peaks and
valleys do not appear and a curve without abrupt changes is obtained. This small variations
detected at high resolution are called DPOAEs fine structures.
Figure 2.6 shows an example of a clinical DPOAE analysis and illustrates its measurement
results. In the figure on the left, it can be seen the sound spectrum measured in a healthy
human ear canal during stimulation of the pure tones f 1 (at 1425 Hz) and f 2 (at 1500 Hz) with
15
CHAPTER 2. HEARING AND SOUND EXPOSURE
Figure 2.5: Development of the traveling waves along the BM in three ases. Graph a) shows the eet in a
dead ohlea. Graph b) shows the eet of a linear ampliation of the outer hair ells. Graph
) ilustrates the eets of the non-linear eet of the outer hair ells. Taken from D.T Kemp
[31℄.
an amplitude of 70 dB SPL both. The rest of the spectral lines that appears are due to the
intermodulation tones produces by the cochlea. The figure on the right shows the result of the
measurements after the presentation of the signals f 1 and f 2 at different frequencies. This is call
a DP-gram. The data of the spectrum of the left figure corresponds to the value shown in the
right figure with an arrow. The shaded portion indicates the noise level during the measurement.
The data is plotted as a function of f 2 .
2.7 Seletion of the Assessment Tehniques for NIHL
A description of the procedures selected in order to analyze the NIHL and the final decision are
presented next.
2.7.1 Seletion of the Audiometry Method for NIHL
In this section the audiometric method is selected from the different audiometric methods presented in Section 2.5. Before deciding the audiometry method, the following requirements are
taken into account:
1. The audiometry method select has to be accurate. The accuracy in this method can be
16
2.7. SELECTION OF THE ASSESSMENT TECHNIQUES FOR NIHL
Figure 2.6: The gure on the left shows the sound spetrum in a healthy human ear anal during stimulation
by two pure tones,
f1
(at
1425 Hz)
and
f2
(at
1500 Hz),
both at 70 dB SPL. The gure on the
righ ilustrates the omplete DP-gram obtained after the measurements.
measured by means of the standard deviation within subjects, which means how vary
the hearing threshold among subjects, being these subjects from a group of otologically
normal persons. In theory all the subjects must have a hearing threshold close to 0 dB
HL, therefore by means of an observation of the mean threshold value averaged across
subjects an estimation of the precision of the audiometry method can be done.
2. The time for each audiometry is in principle not an important factor but a reasonable time
consumption is desired.
3. Conformity with the standard ISO 8253-1 [26] is also desired. Non standardized method
are neglected for this project.
From all the audiometric methods mentioned in Section 2.5 for threshold determination, only
the three standardized methods in ISO 8253-1 [26] are chosen to be analyzed. These are the
Békésy method and the method of Limits using its Ascending and Bracketing versions.
In order to select which one of the three methods mentioned before (Ascending, Bracketing or
Békésy) is more appropriate, an investigation made by Lydolf [34] is used. This author compares different methods in terms of accuracy and time consumption of the measurements. Table
2.1 shows an overview of these parameters. Mean and standard deviation of the used time and
the threshold accuracy for each method are presented. It is observed that the Bracketing method
has the highest mean duration value with 127 seconds and a standard deviation of 12 seconds.
On the other hand in terms of accuracy, the Bracketing method seems to be very precise, although it has a relatively large standard deviation accuracy.
17
CHAPTER 2. HEARING AND SOUND EXPOSURE
Accuracy (dB HL)
Time consumption (seconds)
Ascending
0.8±2
73± 6
Bracketing
0.2±1.9
127±12
Békésy
1.5±1.7
62±12
Table 2.1: Mean and standard deviation of the time used and the HL auray in the threshold determination
of Asending, Braketing and Békésy method. This data is obtained from a investigation of Lydolf
in his Ph.D.Thesis [34℄.
For a simple and a reasonable accurate method, the Bracketing method is chosen. Moreover this
method can be used according to the facilities available for the development of this project.
The sound stimuli used are discrete pure tones at 250, 500,750, 1000,1500, 2000,4000, 6000
and 8000 Hz. The time duration of each pure tone presentation is selected to be 300 ms for all
the pure tones presentations along the frequency range. It was decided to choose this time to
let the subject time enough to hear and to create the response pressing or not pressing when the
sound is respectively heard or not heard [15].
Other parameter is the attenuation step size that indicates how the sounds decrease or increase
in level from one presentation of the signal to the next presentation. This parameter is desired
to be small to give a precise audiometry. If the subject answers that the sound was heard, a
decreasing of 10 dB is performed. However if the subject does not hear the sound an increasing
of 5 dB is performed.
The calculation of the threshold in the standard ISO 8253-1 [26] for the Bracketing method
requires that three runs in series occur at the same level out of a maximum of five runs. Two
ascending-descending runs are chosen in this project instead of three because when using three,
the time consumption of the audiometry with all the frequencies becomes very large.
In the beginning of the audiometry an initial familiarization is carried out to be assure that the
subject has understood how to perform during the test. This familiarizations consist on a pure
tone at an audible level, which decreases until a certain level when the subject is not able to hear
it. If the subject responds correctly to this first descent, the threshold determination goes on. In
this part the attenuation step is the same specified before.
When changing from one frequency to another, the next frequency presented to the subject is
played higher than the level recorded when the push button was not pressed in the last threshold
determination.
2.7.2 Seletion of Otoaousti Emissions for NIHL
Many researchers [38][43][28][42] used OAEs for audiologic evaluation in their investigation
about NIHL.
18
2.8. CHAPTER CONCLUSIONS
Evoked OAEs are frequently used instead of spontaneous OAEs for hearing assessment because
not all the healthy ears produce spontaneous OAEs. Table 2.2 shows the prevalence of the different types of OAEs in human healthy ears which is defined as no cochlear pathology and hearing
threshold of 15 dB HL or better. According with this table, TEOAEs and DPOAEs are the possible evoked OAEs selected for this project.
Type of OAEs
Spontaneous OAEs
Transient Evoked OAEs (TEOAEs)
Stimulus Frequency OAEs (SFOAEs)
Distortion Product OAEs (DPOAEs)
Stimulus
No stimulus required
Click or tone burst
Continuous pure tone
Two pure tones
Prevalence in normal ears
approximately 60%
99%
unknown
99%
Table 2.2: Classiation of the dierent types of OAEs aording to the stimulus applied to exite them and
with their prevalene in normal ears (no ohlear pathology and hearing threshold of
15 dB HL
or
better). The data is obtained from [16℄.
The broad band nature of the TEOAE does not permit a deeper analysis of the hearing due to
its highly non-linearity. Moreover, the frequency analysis of the TEOAEs can not be related to
a specific site of the BM, since the TEOAE level at any given frequency is a result of combined
sources [42]. By contrast, DPOAE allows to associate each DPOAE level with certain areas
of the BM derived from the frequency of the emission signals [42]. Therefore DPOAEs are
preferable to TEOAEs for this project.
DPOAEs allow to into account.
quantify the degree of
DPOAEs are measured at the frequencies 2 f 1 − f 2 , ranging from 635 to 3943 kHz. Two sinusoids with frequencies f 1 and f 2 are selected to be presented to the subject to excite the OAEs.
The distortion product is obtained varying the frequency f 2 from 1001 to 6165 Hz. Fixed levels
of 65 and 45 dB for L1 and L2 respectively and a fixed frequency ratio of f 2 / f 1 = 1.22 are selected for the pure tones. Regarding the resolution of the measurements, eight DPOAE values
per octave band are chosen.
2.8 Chapter Conlusions
Though this chapter, it has been described the hearing assessment in order to detect a possible
hearing damage due to a sound exposure. As stated in the Introduction (Chapter 1), this project
involves the analysis of a specific situation: a headphone sound exposure by MPs.
Headphone sound exposure, as a sound exposure in itself, may leads to NIHL. This NIHL is
diagnosed in this project by means of DPOAE measurements and pure tone audiometry. These
techniques are chosen since the DPOAEs measurements have the potential to indicate a mild
19
CHAPTER 2. HEARING AND SOUND EXPOSURE
NIHL before it is observed audiometrically [43][8].
A summary of the parameters selected in these assessment methods are shown in Appendix A.
In addition, in next chapter a detail description of the sound exposure produced by MPs with
headphones is presented. Furthermore a explanation about how the Lex, To is used to evaluate
this exposure is made.
20
C HAPTER 3
H EADPHONE S OUND E XPOSURE
Headphone sound exposure is a difficult and subjective exposure where many variables play
an important role. This chapter defines in detail how all these variables are considered in this
project for the analysis of a possible NIHL in a population.
3.1 Listening Devies: Musi Players and Headphones
A MP is defined in this project as any kind of device which is able to play music files through
headphones. The sound signal is produced by the MP, however the headphones are the transducers responsible to convert the output signal in sound pressure level.
There are different types of MPs that can be considered for headphone sound exposure, nevertheless, only the following devices are taken into account:
• Laptop computers or desk computers.
• Portable MPs: MP3 players, mobile phones, portable DVD, CD players, portable radios,
pocket computer etc.
• Televisions or games console.
• Hi-Fi equipment, mini-stereo systems, professional stereo systems or similar.
• Mixer table or similar.
All the MP mentioned above must have a common point: They all must have the possibility to
be used with headphones.
Regarding to the headphones, there are five different categories, according to the standard ITUT-1993 [27], that are considered. An explanation of each type can be seen in Appendix B. The
different types of headphones are the following:
• Intra-concha.
• Supra-concha.
• Supra-aural.
21
CHAPTER 3. HEADPHONE SOUND EXPOSURE
• Circum-aural.
• Insert.
In general, headphones do not incorporate acoustic insulation to protect from external noises.
However some headphones in itself provide some attenuation due to their structural properties.
This insulation varies among the different types and among the different models of headphones.
Therefore the exposure level can be modified slightly depending on the type of headphones. This
consideration is not taken into account in this project because it is treated as an uncontrollable
variable because even though some headphones can provide greater attenuation compare to
others, this fact is a consequence of the particular exposure of each person. Thus, it can not be
corrected, because then the real sound exposure will be distorted.
3.2 Listening Soure: Musi
The source of sound exposure through the headphones that is considered for this project is
music.When dealing with noises such as the machinery noisy, the sounds used to be more predictable, so it is easier to analyze. However, in the case of music, the temporal variations, the
frequency content and the average to peak level can vary among different types of music or even
within a same song.
Some investigations [10] have concluded that depending on the type of music, the MP users
adjust the volume of their MP to different levels. It means that the type of music is a parameter
which may affect on the headphone sound exposure. Therefore depending on the music preferences, people can be more or less exposed. Moreover people is not used to listen to only one
type of music. Thus an estimation of the exposure time per each type of music is needed for
each person in order to compare the headphone sound exposure as a function of the music.
An analysis of the headphone sound exposure depending on the type of music is considered
beyond of the objectives of the project.
3.3 Headphone Sound Exposure Evaluation
In this project the headphone sound exposure is calculated as LEX,To which is defined in the
standard ISO 1999 [23] and mentioned in Chapter 2. Equation C.4 shows the expression.
LEX,8h = Leq,Te + 10 log
Te
To
(3.1)
Te is the effective duration that the person is exposed to the noise in hours per day and To corresponds to a reference duration of 8 hours. It is named exposure time. Information about Te
is obtained by means of a questionnaire.
22
3.3. HEADPHONE SOUND EXPOSURE EVALUATION
Leq,Te correspond to the level that a person set in its MP. This is defined as the equivalent continuous sound pressure level which is the steady sound pressure level obtained from an integration
over a period of time of a fluctuating sound pressure level. It is recorded from the sound produced by the headphones. The expression of the Leq,Te can be seen in Equation 3.2. During all
the project the time of integration of all the measurements of Leq,Te is fixed to 60 seconds.
1
Leq,Te = 10 log
Te
Z Te p(t) 2
0
p0
(3.2)
where Te is the time of integration, p(t) is the instantaneous sound pressure level and p0 is the
reference sound pressure which is 20 µPa.
3.3.1 Sound Levels Generated by the Headphones
The measurements of the Leq,60s produced by the headphones of a MP within the ear can be measured with the MANIKIN technique and the MIRE technique stated in the standard ISO 11904-2
and ISO 11904-1 respectively [22][21]. MANIKIN technique uses a standardised artificial ear
including microphones, whereas the MIRE technique uses a probe microphone inserted in the
ears of human subjects.
This last technique can not be used when measuring the sound produced by intra-concha and
insert headphones because the placement of the probe microphone with this type of headphones
is not possible with the correct fit. Furthermore the probe microphones are very delicate, sensitive to environmental conditions and difficult to calibrate [42].
On the other hand, the MANIKIN technique permits to use any king of headphones, and allows
repeated measurements in a short period of time [42]. A disadvantage is the difficulty to use it
outside of the laboratory due to its weight and size. Moreover the pinna of the manikin does
not have the same properties than a human ear, so placement in the ear may not be exactly the
same than in a human being. However this technique is applied in this project since it provides
a simpler and more robust manner of measurement.
The acoustic field produced in the headphone cup can not be compared with a source that creates
a sound field which is propagated in an open space. Therefore a conversion from Leq,60s measured in the ear canal to a field level un-disturbed by the presence of a manikin is needed. This
level is named as free-field related equivalent continuous A-weighted sound pressure level and
it is denoted as LF F,Aeq . Moreover for this project this level is called listening level. Appendix
C explains in detail the procedure to measure and calculate the Leq,60s, the LF F,Aeq and the LEX,8h .
23
CHAPTER 3. HEADPHONE SOUND EXPOSURE
3.4 Headphone Sound Exposure Parameters
The parameters to take into account in order to perform a precise headphone sound exposure
evaluation are explained in this chapter.
3.4.1 Environments
It seems that the acoustic environment around a person may affect the preferred sound level that
a person set in his MP [10].
People use MPs in many different situations. Therefore it is necessary to define precisely the
places and the characteristics of the environments where MPs are used. For instance it can not
be considered the same exposure to utilise a MP with headphones in a very quiet street or a park,
or to use it in a cafeteria. Furthermore in the same environments different background levels are
possible. For example to use a MP while walking in a street can be considered depending on the
street as a very noisy or quiet environment.
The environments that are considered for the headphone sound exposure evaluation are:
• Quiet.
• Moderate.
• Noisy.
3.4.2 Other Sound Exposures
If a NIHL is diagnosed in a person who is exposed to another excessive sound apart from headphone sound exposure by MPs, it is impossible to define which exposure is the responsible of
the hearing damage.
All those persons who do not have any other exposure than the headphone sound exposure
besides not having other complication in his hearing, are considered as individuals who do not
have any previous hearing condition.
For this project a previous hearing condition is presented when a person:
• Suffers from a cold or has taken any medicine or another type of drugs that might influence
on his hearing.
• Has very often hearing problems as infections, ear noises, tinnitus, drainage etc.
• Has a know history of hearing damage diagnosed by a medical doctor.
• Has been exposed to impulsive or loud sounds for long periods of time without hearing
protectors i.e. explosions, fireworks, shootings etc.
24
3.5. CHAPTER CONCLUSIONS
• Has been working in a job were the use of headphones was needed for at least two years
full time.
• Has been working in a noisy environment for at least two years full time without using
hearing protectors.
• Is a musician or is professionally involved.
• Used to play an instrument or attend to concerts/discos very often without using hearing
protectors.
3.4.3 Gender and Age Dierenes
Among MP with headphones users, there are people of different ages. This project investigates
the age range of the potential users of MPs with headphones in a population. However whether
the age within the potential users is a parameter that affects the headphone sound exposure is
not an issue analysed in this project.
In addition, whether a gender difference exist when assessing noise exposure through the headphones is not an issue to be investigated.
3.5 Chapter Conlusions
In the headphone sound exposure evaluation carried out for this project, different types of headphones and MPs are selected to be analysed when music is played. Moreover quiet, moderate
and noisy are the three environments considered.
In order to assess the influence that MPs with headphones produces on the hearing, a selection
of MP users according to their LEX,8h and their previous hearing condition is made.
Next chapter explains the classification of subjects in two groups, defined as control and target
population:
• The control population is a group of persons that do not present any previous hearing
condition besides not being potential MP users.
• The target population is formed by persons who do not present any previous hearing
condition besides listening to MP with headphones for long periods of time and at high
volume settings.
25
C HAPTER 4
P ILOT T EST
The pilot test is performed in order to select the subjects that belong to control and target population for the purposes of this project. First a preliminary survey is described, and afterwards
the pilot test design is explained.
4.1 Preliminary Survey
The purpose of this part is to define the age of MPs users that utilize headphones. This survey
is conducted to 80 subjects randomly selected asking them the next question: "Do you listen to
music using MP with headphones?" The important parameters that are taken into account are
the age, the gender, and the answer to the question formulated.
The data collected is shown in Table 4.1. From these results, it can be seen that the highest
percentage of MP with headphones users is 83,3% which corresponds to an age ranged from
19-30. In consequence it is stated that the age range of the target and control population that
will be tested in this project has to be from 19 to 30 years old.
Number of subjects
Women
Men
10
10
10
10
10
10
10
10
Age Range
19-30
31-38
42-48
51-57
% of persons who use
MP with headphones
83.3
63.3
33.3
26.6
% of persons who do not
use MP with headphones
16.6
36.6
66.6
73.3
Table 4.1: Perentage of the population that use or do not use MP with headphones.
4.2 Pilot Test Design
In this section the previous hearing condition, the listening habits and the LEX,8h are studied. It
has to be noticed that the subjects who performed the pilot test are not the same subjects who
participated in the preliminary survey. The only requirement for the new subjects is to be ranged
in age from 19 to 30, as it was concluded in the results of the preliminary survey.
27
CHAPTER 4. PILOT TEST
4.2.1 Goals of the Pilot Test
The goals of the pilot test are:
• To identify two groups of subjects: target and control population
• To define the most typical places where MPs are used.
This is done by means of
• Analysis of the previous hearing condition and the environments where MPs with headphones are used. The subjects could select among the next environments: Bus, train, bike,
motor-bike, car, street, home, university or at work. This is done with a questionnaire.
• An evaluation of the headphone sound exposure of the subjects testing the listening level
and the exposure time.
In addition, some other parameters are investigated in order to verify some of the decisions that
were taken in the pilot test design. These are:
• The use of MP with headphones among the next devices: Portable MP, computer, television/DVD, HIFI equipment or mixer table.
• The type of headphones preferred by users among the next types: Intra-concha, supraconcha or circum-aural headphones.
4.2.2 Method
It is decided to carry out this pilot test in Aalborg university offices which is a quiet environment
where the background noise can not vary suddenly from low to high levels. The instructions
given to the subjects are shown in Appendix E.
This pilot test is divided in two steps:
1. It consists of testing the volume listening preferences of the subjects when using MP with
headphones. The purpose is to obtain the sound exposure level of each subject. In order
to do that, two different types of headphones and one portable MP are presented to every
subject. They can be seen in Table 4.2. Each person must select the type of headphones
that would choose in case of listening to music with headphones. Then, the task of the
subject will be to adjust the preferred volume control of one from seven different music
samples of 60 seconds that are stored in the portable MP. Table 4.3 shows title, author and
type of the music samples. A normalization of all the music samples is done to assure that
they all have the same energy content. The music samples are popular and well known
songs in order to allow the subjects to focus on the volume selection instead of another
not important details i.e. the lyrics or the rhythm of the music.
28
4.2. PILOT TEST DESIGN
Appendix D describes the selection of the portable MP and the headphones as well as the
music samples used. The normalization performed in the music samples is also detailed
in this appendix.
Device
Headphone
Headphone
MP
Model
Sony MDR
Creative MuVo V200
Creative MuVo V200
LAB-Nr
2157-41
2157-41
Table 4.2: Devies used in the pilot test
2. It is based on a paper questionnaire inspired by the one used by Mattila & Zacharov [35].
This questionnaire is formulated in order to identify those subjects who do not have any
previous hearing condition and to get some information related to their listening habits.
The entire questionnaire can be seen in the Appendix F.
Music Author
The Beetles
Milk Inc.
Fugees
Woody Allen
Shakira
U2
Vivaldi
Music Title
Help
In my eyes
Ready or not
Come On and Stomp
Hips dont’t lie
Sunday Bloody Sunday
The Four Seasons
Music Type
Pop
Dance
Hip-hop
Jazz
Pop
Rock
Classical
Table 4.3: Musi samples presented to the subjets. The duration of all the musi samples is 60 seonds
4.2.3 Evaluation and Results
A total of 61 persons, 40 males and 21 females, ranged in aged from 21 to 33 (median=23) took
part of this study. Appendix G shows the results in detail.
Previous Hearing Condition
Firstly, the subjects that do not present any previous hearing condition are selected and the rest
are rejected.
Regarding to the results collected from the questionnaire, any significant problem related to
hearing infections or medicine that might have influence on the hearing was revealed by the
subjects. Five candidates showed a diagnosed hearing problem and one of them was suffering
a cold, therefore these six subjects, that correspond to a 9.83% of the tested population were
29
CHAPTER 4. PILOT TEST
rejected.
There was not relevant case affected by impulsive or loud sounds as explosions, shootings or
firearms. The most common cases correspond to persons exposed to fireworks once or twice
per year. In addition there was a 8% of subjects that declared that they have been working in a
noisy environments i.e. machinery noises in a factory. However it was during short periods of
time and in the most of cases hearing protectors were used.
A 32% of the subjects play some musical instrument as piano, guitar, drums, saxophone or electrical guitar. The 11.76% of these subjects play in a band in their free time and 88,24% play at
home. There is only one subject who used to play around 80 hours per month, therefore this
participant is rejected. The rest of the subjects are accepted because any of them was professionally involved in music and the exposure time was less than 32 hours.
A 82,69% of the subjects assists to discos or pop/rock concerts however since any subject attends regularly and the average is 2 times per month, then it is considered that any subject is
severely affected.
Thus, it can be concluded that 54 subjects, that corresponds to 88.52% of the population, are
considered do not present any previous hearing condition. From these subjects the headphone
sound exposure is analyzed.
Headphone Sound Exposure
The next step is to calculate the headphone sound exposure to identify the subjects of the target
and control population using the data collected from the control level setting and the time exposure marked in the questionnaire. This is done by means of the calculation of the LEX,8h as it
was explained in Section 3.3 in Chapter 3.
The levels that correspond to each control level setting are measured beforehand, so they were
known before the performance of the pilot test. This can be seen in Appendix E.
The LEX,8h of all the subject can be seen in Table G.1 in Appendix G. Figure G.1 presents
the LEX,8h sorted from the minimum to the maximum LEX,8h . There are 11 subjects that are not
MP users, therefore the exposure time is zero and consequently the corresponding LEX,8h is zero.
This data is evaluated according to the Danish legislation [5] which states that no person may
be subjected to noise with a level higher than 85 dBA in an eight hour working day. Therefore
those persons who are exposed to levels above 85 dBA are considered subjects of the target population. On the other hand, all the subjects who performed a LEX,8h of less than 75 dBA are
considered control population.
30
4.2. PILOT TEST DESIGN
110
100
90
80
LEX 8h [dB SPL]
70
60
50
40
30
20
10
0
Figure 4.1:
LEX,8h
1
5
10
15
20
25
30
35
Reference number of the subjects
40
alulated from the data olleted in the pilot test. The
from the minimum to the maximum
45
LEX,8h
50
54
of the subjets is sorted
LEX,8h
The difference between the LEX,8h in the population would be desired to be bigger, but due to
lack of subjects interested to participate in the complete investigation, this range was reduced to
be only 10 dB difference between populations.
Parameters Analysis
This last step evaluates the parameters mentioned in Section 4.2.1 using the data collected from
the questionnaire.
Regarding to the environments where MPs with headphones are used, the subjects were asked
about how often they used to listen to their MPs (rarely, often or very often) and the type of
environment that corresponds to the place selected (quiet, moderate or noisy).
The analysis is made according to a 3-stage analysis system that is invented. A point scale from
1 to 3 corresponding 1 to the weakest stage and 3 to the strongest as it can be seen in the questionnaire is used. Figure 4.2 shows the results plot as a function of the number of points get
in each environment. It can be seen that the two environments that get the highest number of
points were bus and street.
A representation of the number of MPs users for each type of MP devices is shown in Figure
4.3. The participants display a high use of MP devices when headphones are used, especially
31
CHAPTER 4. PILOT TEST
150
Number of Points
100
50
k
or
w
ity
iv
e
rs
m
e
un
Types of Environments
ho
in
tra
r
ca
bi
ke
m
ot
or
bi
ke
et
st
re
bu
s
0
Figure 4.2: Environments where MPs with headphones are used. Eah subjet ould hoose more than one
environment.
portable MP and personal computers or laptops are the MP most used by the population tested.
Therefore it is consistent with the decision taken of using a portable MP as reference device in
order to measure the headphone sound exposure.
Regarding the type of headphones that the population uses, it can be concluded that 71.11%
use intra-concha headphones, 8.89% use supra-concha headphones and 20% use circum-aural
headphones. Therefore the headphones selected in Appendix D for this pilot test are according
to the preferences of the people that use headphones for listening to MPs.
4.3 Chapter Conlusions
When calculating the headphone sound exposure for each subject, the same exposure level was
considered for the total exposure time. This assumption is only an approximation because it is
not proved that the MP users listen to the music at the same level all the time.
The users may modify the control volume position of the MP depending on the environment
where they are listening to music. Due to this, another headphone sound exposure evaluation
taken into account different environments is performed in a listening test that is explained in
next chapter. Bus and street, which are the most common places to use the MP as it was concluded in this pilot test, are used as references to evaluate the LEX,8h in a quiet/moderate and
noisy environment respectively.
Furthermore, subjects did not use their MPs or their headphones in this pilot test. This may
32
4.3. CHAPTER CONCLUSIONS
Clasification of music player users
50
45
40
Number of subjects
35
30
25
20
15
10
5
0
PMP users
PC users
TV/DVD users
Type of device
HIFI users
Mixer users
Figure 4.3: Use of MPs with headphones as a funtion of the users preferenes.
Eah subjet had the
possibility to selet more than one devie
change the control level setting somehow for a lower or higher setting. Therefore it is asked to
the subjects to bring their own MP, headphones and music in the listening test. Thus, a more
accurate evaluation of the real headphone sound exposure can be made.
From the total of persons that took part of this pilot test, 18 subjects (28%) are selected as control
population and 26 subjects (33%)as target population. These subjects are called to participate in
a hearing assessment and a headphone sound exposure evaluation, which is carry out in different
environments.
33
C HAPTER 5
L ISTENING T EST
This chapter describes the design of the listening test. First the method and the goals of the
listening test are described. Then, a detail description of each part is explained. The results of
this listening test will be given in detail in Appendix. K.
5.1 Goals of the Listening Test
The goals of the listening test are:
• To measure the listening level, LF F,Aeq , in three environments: very quiet, quiet/moderate
and noisy for the control and the target population.
• To calculate the LEX,8h for both populations taken into account two environments: quiet,
quiet/moderate and noisy.
• To test the hearing thresholds and the DPOAEs for both populations.
5.2 Listening Test Design
The test is made as it explained in the following lines:
First the subject is conducted to a room where an instruction paper is given. It can be seen in
Appendix H. Then the subject must fill out a short questionnaire. After this, the subject is conducted to an audiometer cabin where DPOAEs measurements and a pure tone audiometry are
performed. Then, a break is given to the subject to be ensure that he does not get tired and lose
focus. After the break, the subject is conducted to a listening room. A control volume adjustment of his MP using his headphones and music is performed in different environments several
times. Between each volume adjustment the subject come out of the room during 1 or 2 minutes.
The paper instructions given to the subjects can be seen in Appendix H. In addition, the conducted guidelines followed by the instructor can be found in Appendix I.
35
CHAPTER 5. LISTENING TEST
5.3 Listening Test Subjets
To perform this listening test, 26 and 18 subjects selected from the pilot test as target and control
population respectively are invited to participate in this second part of the investigation. From
these subjects, 12 and 10 subjects of the target and control population respectively agreed to
participate.
Table K.2 and K.1 in Appendix K shows reference number, age and sex of all the subjects of
control and target population who participate in the listening test respectively.
5.4 First Part: Listening Test Questionnaire
The first task is to fill out a questionnaire. This questionnaire can be seen in Appendix J.
The objective is to estimate the exposure time when using headphones for listening to music
with a MP in a quiet/moderate and noisy environment for all the subjects. The results can be
seen in Figure K.2 of Appendix K.
In this questionnaire, as well as in the pilot test, the subjects who suffer from a cold or take any
medicine or another type of drugs are rejected. Any subject was in this situation, therefore no
subject was rejected.
Moreover there are some specific questions asked to have a better overview of the hearing state
of the subjects. The answers given by the subjects can be seen in Figure K.9 in Appendix K.
5.5 Seond Part: Hearing Assessment
An audiometry test and DPOAEs are carried out. Appendix L describes these measurements
and presents the results.
From the target population, two subjects are neglected after the performance of these two tests.
Subject EDR is neglected because a deep hearing loss at high frequency in his right hear was
detected. This audiometry was repeated twice and the results obtained were approximately
the same. The results of his audiometry can be observed in Table L.2 in Appendix L. This
hearing loss is considered to be produced by other reasons not related to the use of MPs with
headphones, therefore this subject is neglected. Moreover, there was a problem conducting the
DPOAE measurement to the subject JLS. It was seen that after the DPOAE measurement of the
right ear, the measurement system was blocked. It was due to an accumulation of wax that was
36
5.6. THIRD PART: MP VOLUME ADJUSTMENT
stuck in the microphone probe of the measurement system. Then, the wax was removed and the
measurement finished. Observing his results, it can be seen that there is a strong asymmetrical
difference between the right and left ear of both tests. After this examination, it is decided not
to take into account this subject because there are some signs that this subject had an excessive
amount of wax in his right ear. His audiometry and DPOAE measurements can be seen in Table
L.2 and Figure L.17 respectively in Appendix L.
5.6 Third Part: MP Volume Adjustment
One of the goals of this listening test is to analyze the LEX,8h of both population. This parameter
is calculated following the same procedure used in the pilot test. Therefore, first the subject
must adjust the volume of his MP, then LFF,Aeq is measured, and finally with this level and the
exposure time,LEX,8h is obtained.
The main considerations in this listening test are:
• The subjects are asked to bring their own MP, headphones and music. Since not all the
MPs are portable, computers and MPs easily transported are recommended. Moreover the
devices used in the pilot test are available just in case that the subjects can not bring their
own devices.
• Bus and street, which are the most common places to use the MP as it was concluded
in the pilot test in Chapter 4, are used as references environments to evaluate the LEX,8h
in a quiet/moderate and noisy environment respectively. Moreover the subject is asked
to adjust the MP when there is not environment being simulated. The environments are
presented to the subjects randomly. Furthermore, two repetitions of each environment are
simulated. Appendix M explains in detail the listening test setup and the simulations of
the environments.
• For the calculation of the LEX,8h , only the quiet/moderate and the noisy environment are
considered since a very quiet situation is considered as an unrealistic environment.
• A familiarisation process is performed before starting this part of the listening test. It
consist of an adjustment in one of the environments. This adjustment is not taken into
account for the results.
The LEX,8h obtained for all the subjects is shown in Figure 5.1. Figures and tables of the exposure
level and the exposure time can be found in Appendix K.
37
CHAPTER 5. LISTENING TEST
5.7 Chapter Conlusions
The data from 10 subjects of the control and 10 subjects of the target population are collected to
be analyzed in next chapter.
From the pure tone audiometry and DPOAE measurements, 10 hearing threshold levels (HL)
and 22 DPOAE levels are obtained respectively per subject and per ear. These values are according to the frequencies selected in section 2.7 in Chapter 2.
From the headphone sound exposure evaluation, for each subject, two exposure levels (LEX,8h )
and three listening levels (LF F,Aeq ) are calculated. The exposure levels corresponds to a quiet/moderate and a noisy environment. Moreover, the listening levels corresponds to a very quiet,
quiet/moderate and noisy environment.
Control Population
Lex,8h [dB SPL]
100
90
80
70
60
VLL
DVD
HDA
MCK
ANT
DNL
BTZ
LRM
IRS
HCR
Quiet/moderate Environment
Noisy Environment
Target Population
Lex,8h [dB SPL]
100
90
80
70
60
Figure 5.1:
LEX,8h
LRI
CLS
ERC
SMN
YSN
NCL
Subjects
SUS
OLV
JON
CLA
alulated for the subjets of the target and ontrol population aording to the data
obtained in the listening test for a quiet/moderate and a noisy environment.
38
C HAPTER 6
L ISTENING T EST R ESULTS
This chapter contains an evaluation of the conducted listening test in order to analyze the data
collected using an statistical tool.
6.1 Listening Test Analysis
There are four parameters tested: exposure level (LEX,8h ), listening level (LFF,Aeq ), hearing
threshold level (HL) and DPOAE level. Moreover, there are some factors that are controlled
in the listening test. Each factor contains a number of treatments. All factors and treatments
used in this analysis are list below:
• Factor: Populations
– Treatment: Control population
– Treatment: Target population
• Factor: Ears
– Treatment: Left ear control population (LC)
– Treatment: Right ear target population (RC)
– Treatment: Left ear control population (LT)
– Treatment: Right ear target population (RT)
• Factor: Environments
– very quiet
– quiet/moderate
– noisy
• Factor: Frequencies
– Treatment: fi
where:
i=1,2,...,10 for the HL statistical analysis.
i=1,2,...,22 for the DPOAE statistical analysis.
39
CHAPTER 6. LISTENING TEST RESULTS
In order to perform this statistical study an ANalysis Of VAriance test (ANOVA) and an Independent Sample T-test are applied using the Program SPSS 14.0. For further reading on the
method, reference is made to [36]. In Appendix N and Appendix O a further explanation about
the ANOVA and the T-test analysis is described.
6.2 Analysis of Listening Level
The aim of this section is to determine if the listening level differs among the populations and
the environments.
In this study, the parameter analyzed is the LF F,Aeq . The factors and the treatments defined are
the followings:
• Factor A: Population
– Treatment µA1 : Control population
– Treatment µA2 : Target population
• Factor B: Environments
– Treatment µB1 : Very quiet environment
– Treatment µB2 : Quiet/moderate environment
– Treatment µB3 : Noisy environment
All the subjects, regardless of the type of population that they belong to, are introduced to the
three environments simulated.
The analysis of the listening level is done to answer the next questions:
• Do control and target population differ significantly on their performance on the LFF,Aeq ?
• Second, is there any significant difference in listening level among very quiet, quiet/moderate and noisy environment?
• Finally, how do the two factors: populations and the environments interact in their effect
on LFF,Aeq ?
6.2.1 Method
Since this test has two factors, it is analyzed using a two-way ANOVA test. Therefore, it is
possible to evaluate the influence of the two factors as well as the influence of a possible interaction between them. In order to evaluate this statistical significance the next hypotheses are
formulated:
40
6.2. ANALYSIS OF LISTENING LEVEL
• For factor A: Populations.
– The null hypothesis, H0,A : It states that there is not significant difference between
the LFF,Aeq means of the two populations. It is noted as:
H0,A : µA1 = µA2 .
– The alternative hypothesis, H1,A : It states that there is a significant difference between the LFF,Aeq means of the two populations.
H1,A : µA1 6= µA2 .
• For factor B: Environments.
– The null hypothesis, H0,B : It states that there are not significant differences among
LFF,Aeq means of the three environments.
H0,B : µB1 = µB2 = µB3 .
– The alternative hypothesis, H1 , B: It states that at least one of the three LFF,Aeq means
is different from the others.
• For interaction between factors population and environments.
– The null hypothesis, H0,AB : It states that the interaction between factor populations
and environments has a significant effect on LFF,Aeq .
– The alternative hypothesis, H1,AB : It states that the interaction between factor populations and environments has not a significant effect on LFF,Aeq .
6.2.2 Results
Table 6.1 presents the LFF,Aeq means across subjects for the treatments and factors defined in
this section. Moreover the two-way ANOVA test results are presented in Table 6.2. Figure
6.1 illustrates the Confidence Interval (IC) and the estimated LFF,Aeq means for the two factors:
populations (left) and environments (right).
Factor B
Environment
very quiet
quite/moderate
noisy
Mean for population (dB)
Mean LFF,Aeq (dB)
Control Target
87.82
97.29
94.11
98.83
97.52
101.39
93.15
99.17
Standard Deviation
Control
Target
9.99
5.75
5.10
4.47
3.38
4.42
Number of Subjects
Control
Target
10
10
10
10
10
10
Table 6.1: Data olleted from the listening test for the
41
LFF,Aeq
analysis.
Mean for
environment (dB)
92.96
96.98
99.76
CHAPTER 6. LISTENING TEST RESULTS
Factor A: Population
Factor B: Environment
Interaction AxB
Error
Total
Sum of Squares
388,524
483,794
72,408
1830,744
571229,936
Table 6.2: Results of the
LFF,Aeq
Degrees of Freedom
1,000
2,000
2,000
54,000
60,000
Mean Squares
388,524
241,897
36,204
33,903
F-value
11,460
7,135
1,068
p-value
0,001
0,002
0,351
analysis using two-way ANOVA statistial method.
Figure 6.1: Representation of the listening level as a funtion of the population (left) and the environment
(right). The Null hypothesis an be rejeted if the varianes lines do not overlap in the horizontal
diretion.
6.2.3 Conlusions
In this project the level of significance used is equal to 0.05. Then, to reject a null hypothesis,
the p-value in the ANOVA table must be lower than 0.05.
From Table 6.1 it can be seen that the two factors present differences among the LF F,Aeq means.
Furthermore, it is shown in Table 6.2 that the p-value in both cases is lower than 0.05. Then, the
null hypotheses H0,A and H0,B can be rejected. This means that there is a significant difference
between the performance of both populations. In addition, it also can be concluded that the
listening level varies significantly among the three environments.
Since the p-value from factor population and environment interaction is higher than 0.05. It can
be concluded that the interaction between these factors has not significant effect on the listening
level performance.
The same conclusions can be observed from Figure 6.1. If the variance lines that represents
42
6.3. ANALYSIS OF EXPOSURE LEVEL
the IC from the estimated LFF,Aeq does not overlap in the horizontal direction, two means are
considered significantly different. In this case, the variance lines does not overlap, then the populations and the environments means are considered significantly different.
In order to analyze in detail the effect of LFF,Aeq among the three environments an Independent
Sample T-test is performed. The T-test design and the results obtained are presented in Appendix O.
From the T-test results shown in Tables O.1, O.2 and O.3, the next conclusions are made. There
is not a significant difference on LF F,Aeq performance between very quiet and quiet environment
and between quiet/moderate and noisy environment. However, there exists a difference between
very quiet and noisy environment LF F,Aeq performance.
6.3 Analysis of Exposure Level
The goal of this section is to determine if the LEX,8h of the control and the target population
would differ among the environments.
This study is similar to the sound exposure level analysis described in Section 6.2. However, in
this case the parameter analyzed is the LEX,8h . The factors and the treatments are:
• Factor A: Population
– Treatment µA1 : Control population
– Treatment µA2 : Target population
• Factor B: Environment
– Treatment µB1 : Quiet/moderate environment
– Treatment µB2 : Noisy environment
Those subjects who belong either to the control or target population are introduced to all the
treatments defined by the factor environments. Notice that now factor B has two treatments.
Since the LEX,8h of the subjects that do not use MP with headphones is equal to zero. These
subjects are not taken in account in this study.
The analysis of the exposure level is done to answer the next questions:
• First, do control and target population differ significantly on their performance on LEX,8h ?
• Second, is there any significant difference in LEX,8h between quiet/moderate and noisy
environment?
43
CHAPTER 6. LISTENING TEST RESULTS
• Finally, how do the two factors: populations and the environments interact in their effect
on LEX,8h ?
6.3.1 Method
To evaluate the influence of factor populations, environments and the interaction between them
a two-way ANOVA analysis is applied. Thus, the hypotheses formulated are:
• For factor A: Populations.
– The null hypothesis, H0,A : It states that there is not significant difference between
the LEX,8h means of the two populations. It is noted as:
H0,A : µA1 = µA2 .
– The alternative hypothesis, H1,A : It states that there is a significant difference between the LEX,8h means of the two populations.
H1,A : µA1 6= µA2 .
• For factor B: Environments.
– The null hypothesis, H0,B: It states that there is not a significant difference between
the LEX,8h means of the two environments.
H0,B : µB1 = µB2 .
– The alternative hypothesis, H1,B : It states that there is a significant difference between the LEX,8h means of the two environments.
H1,B : µB1 6= µB2 .
• For interaction between factor populations and environments.
– The null hypothesis, H0,AB : It states that the interaction between factor populations
and environments has a significant effect on LEX,8h .
– The alternative hypothesis, H1 : It states that the interaction between factor populations and environments has not a significant effect on LEX,8h .
6.3.2 Results
Table 6.3 shows the mean LEX,8h for the treatments and factors considered. In addition, in Table
6.4 the results from the ANOVA analysis are presented.
A comparison of the mean listening levels and the IC of the two treatments of factor population(left) and environment (right) can be shown in Figure 6.2.
44
6.4. ANALYSIS OF HEARING THRESHOLDS AND DPOAES
Factor B
Environment
quite/moderate
noisy
Mean for population (dB)
Mean LEX,8h (dB)
Control Target
82.86
89.68
78.31
88.33
77.58
89.00
Standard Deviation
Control
Target
6.17
3.58
7.76
9.55
Table 6.3: Data olleted from the listening test for the
LEX,8h
Number of Subjects
Control
Target
5
10
5
10
Mean for
environments (dB)
86.27
83.32
analysis. The subjets that do not use MP
are nor inluded in this study.
Factor A: Population
Factor B: Environment
Interaction AxB
Error
Total
Sum of Squares
432,944
0,872
124,639
1539,139
254214,503
Table 6.4: Results of the
LEX8h
Degrees of Freedom
1,000
1,000
1,000
30,000
34,000
Mean Squares
432,944
0,872
124,639
51,305
F-value
8,439
0,017
2,429
p-value
0,007
0,897
0,130
analysis using two-way ANOVA statistial method
6.3.3 Conlusions
Regarding the data exposed in Table 6.3 and Figure 6.2 (left), it can be seen that the LEX8h means
of factor populations are significantly different. Moreover the p-value of this factor, shown in
Table 6.4, is equal to 0.007. This means that H0 , A can be rejected. Therefore the LEX8h varies
depending on the population.
The means of the factor environments are also different. However Figure 6.2 (right) shows that
the IC are almost overlapped. Since the p-value is significantly higher than 0.05, H0 , B is not
rejected. Then it can be concluded that LEX,8h do not depend on the environment.
From Table 6.4, it can be appreciate that the interaction effect between factor environments and
populations is not present since the p-value is equal to 0.130.
6.4 Analysis of Hearing Thresholds and DPOAEs
In this section the data obtained from the audiometric tests and the DPOAE measurements is
analyzed. Since the procedure followed is similar, both analysis are explained in the following
lines.
The purposes of this section are:
• To analyze if the HL differs for both populations and for the 10 frequencies tested.
• To analyze if the DPOAE level differs for both populations and for the 22 frequencies
45
CHAPTER 6. LISTENING TEST RESULTS
Figure 6.2: Representation of
LEX,8h
in funtion of the population (left) and the environment (right). The
null hypothesis an be rejeted if the varianes lines do not overlap in the horizontal diretion.
tested.
There are two factors: ears and frequencies. The parameter analyzed is HL or DPOAE level
depending on the test. The treatments contained by the two factors are listed below:
• Factor A: Ears
– Treatment µA1 : LC
– Treatment µA2 : RC
– Treatment µA3 : LT
– Treatment µA4 : RT
• Factor B: Frequencies
– Treatment µBi : fi
where:
i=1,2,...,10 for the HL statistical analysis.
i=1,2,...,22 for the DPOAE statistical analysis.
It should be noticed that the range of frequencies tested is different in the audiometric test than
in the DPOAE measurements.
Figures 6.3 and 6.4 present the HL mean and the standard deviation of the two ears of both
populations along the frequencies.
46
6.4. ANALYSIS OF HEARING THRESHOLDS AND DPOAES
−15
Mean Right Ear Control Population
Mean Left Ear Control Population
−10
Threshold [dB HL]
−5
0
5
10
15
20
25
250
500
750
1k
1.5 k
2k
Frequency [Hz]
3k
4k
6k
8k
Figure 6.3: Representation of the HL mean and the standard deviation of left and right ear of ontrol
population.
The data collected from the DPOAE test is presented in Figures 6.5 and 6.6. It represents
the DPOAE levels of left and right ear of both populations and for the frequencies defined as
√
f 1. f 2.
All these data is calculated by averaging across the ears of 10 subjects of control population and
10 subjects of target population. All these values are shown in Table P.1 and P.2 of Appendix P.
From an analysis of the data presented above, the next questions can be stated:
• First, do LC, RC, LT and RT ears differ significantly on their HL performance? Do they
differ on DPOAE level performance?.
• Second, is there any significant difference in HL among the frequencies tested? Is there
any difference in DPOAE level?
• Third, how do the two factors: ears and frequencies interact in their effect on HL? How
do they interact in their effect on DPOAE level?
6.4.1 Method
It is desired to analyze the performance of the parameter HL or DPOAE level on the factors ears
and frequencies. Thus, a two-way ANOVA test is applied individually for HL and DPOAE level
data.
In next lines the hypotheses that corresponds to the hearing thresholds analysis are enunciated:
47
CHAPTER 6. LISTENING TEST RESULTS
−15
Mean Right Ear Target Population
Mean Left Ear Target Population
−10
Threshold [dB HL]
−5
0
5
10
15
20
25
250
500
750
1k
1.5 k
2k
Frequency [Hz]
3k
4k
6k
8k
Figure 6.4: Representation of the HL mean and the standard deviation of left and right ear of target popu-
lation.
• For factor A: Ears.
– The null hypothesis, H0,A : It states that there are not significant differences among
the HL means of the four ears. It is noted as:
H0,A : µA1 = µA2 = µA3 = µA4 .
– The alternative hypothesis, H1,A : It states that at least one of the ears means is different from the others.
• For factor B: Frequencies.
– The null hypothesis, H0,B : It states that there are not significant differences among
the HL means of the ten frequencies.
H0,B : µB1 = µB2 = µBi .
where i is the number of frequencies.
– The alternative hypothesis, H1 , B: It states that at least one of the HL means of the
ten frequencies is different from the others.
• For interaction between factor ears and frequencies.
– The null hypothesis, H0,AB : It states that the interaction between factor ears and
frequencies has a significant effect on HL.
– The alternative hypothesis, H1,AB : It states that the interaction between factor ears
and frequencies has not a significant effect on HL.
Since the hypotheses of DPOAE analysis are similar, they are not described in this report.
48
6.4. ANALYSIS OF HEARING THRESHOLDS AND DPOAES
30
Mean DPOAE Left Ear Control Population
Mean DPOAE Right Ear Control Population
Mean Noise DPOAE Left Ear Control Population
Mean Noise DPOAE Right Ear Control Population
25
20
DPOAE [dB SPL]
15
10
5
0
−5
−10
−15
−20
1000
1500
2000
3000
Frequency [Hz]: geometric mean
4000
6000
Figure 6.5: Representation of the DPOAE mean and the standard deviation for left and right ear of ontrol
population.
6.4.2 Results
In the following lines the results obtained from the two-way ANOVA test are presented for
hearing thresholds and DPOAE analysis.
Hearing Thresholds results
In Table P.1 of Appendix P it is shown the HL means for the four ears and for the ten frequencies
tested. Moreover, Table 6.5 presents the two way ANOVA test results obtained. The HL means
and the IC of the factors ears and frequencies are shown in Figure 6.7.
Factor A: population
Factor B: frequency
Interaction AxB
Error
Total
Sum of Squares
934,000
2942,750
1047,250
13395,000
21800,000
Degrees of Freedom
3,000
9,000
27,000
360,000
400,000
Mean Squares
311,333
326,972
38,787
37,208
F-value
8,367
8,788
1,042
Table 6.5: Results of the HL analysis using two-way ANOVA statistial method
49
p-value
0,000
0,000
0,410
CHAPTER 6. LISTENING TEST RESULTS
30
Mean DPOAE Left Ear Target Population
Mean DPOAE Right Ear Target Population
Mean Noise DPOAE Left Ear Target Population
Mean Noise DPOAE Right Ear Target Population
25
20
15
DPOAE [dB SPL]
10
5
0
−5
−10
−15
−20
1000
1500
2000
3000
Frequency [Hz]: geometric mean
4000
6000
Figure 6.6: Representation of the DPOAE mean and the standard deviation for left and right ear of target
population.
DPOAE results
Table P.2 of Appendix P shows the DPOAE means of the four populations and for the 22 frequencies tested. In addition, Table 6.6 presents the two way ANOVA test results obtained. A
representation of the DPOAE means and the IC of the two factors analyzed is presented in
Figure 6.8.
Factor A: ears
Factor B: frequencies
Interaction AxB
Error
Total
Sum of Squares
1387,206
5359,802
1088,462
31026,891
47890,850
Degrees of Freedom
3,000
21,000
63,000
792,000
880,000
Mean Squares
462,402
255,229
17,277
39,175
F-value
11,803
6,515
0,441
p-value
0,000
0,000
1,000
Table 6.6: Results of DPOAE level analysis using two-way ANOVA statistial method
6.4.3 Conlusions
This section presents the conclusions obtained from the HL and DPOAE analysis.
HL Conlusions
From Table 6.5 it can be seen that the p-value for factor ears is less than 0.05. Then, the null
hypothesis H0,A can be rejected. This means that at least one of the four treatments of factor
50
6.4. ANALYSIS OF HEARING THRESHOLDS AND DPOAES
Figure 6.7: Representation of HL mean in funtion of the ears of both populations (left) and the frequenies
(right)
ears performs different from the rest. There is at least a significant difference among the four
treatments analyzed.
In case of factor frequencies, the null hypothesis H0,B can also be rejected as the p-value is lower
than 0.05. Therefore, a significant difference among frequencies can also be found.
Nevertheless, the null hypothesis H0,AB can not be rejected. Then it can be concluded that there
is not interaction effect between factor ears and frequencies.
Because H0,A and H0,B are rejected, it is known that at least a treatment performs different than
the others. In order to study in detail the differences among the treatments of the factor ears and
the factor frequencies, a T-test is performed for each factor:
• The T-test results for the factor ears are presented in Table O.4 of Appendix O. It can
be concluded that there is a dependence on HL between the left and the right ear of the
control population. This dependence is also found in the target population. There is not
dependence in the HL performance between left and right ears of different populations.
For instance, LC performs different than LT and RT.
• The T-test results for the factor frequencies are shown in Table O.5 of Appendix O. It
can be appreciated a significant difference on HL performance between 6000 and 8000 Hz
frequencies and the others.
51
CHAPTER 6. LISTENING TEST RESULTS
Figure 6.8: Representation of DPOAE level mean in funtion of the ears of both populations (left) and the
frequenies (right).
DPOAE Conlusions
The p-values shown in Table 6.6 for factor ears and frequencies are less than 0.05. This means
that there is at least a difference among the treatments of factor ears. This is also the same situation for factor frequencies. However, the interaction between them is not significant since the
p-value is higher than 0.05.
In order to study in detail the differences among the treatments of the factor ears and factor
frequencies, a T-test is performed for each factor:
• The T-test results calculated to analyze the differences of DPOAE level performance
among the treatments of the factor ears, are shown in table O.6 of appendix O. A significant dependence between the left and the right ear of both populations is also appreciated.
• The T-test results for the factor frequencies are shown in Table O.7 of Appendix O. It can
be appreciated a significant difference on DPOAE level performance in a range between
3626.17 and 5584.93 Hz, compared to the rest.
6.5 Chapter Conlusion
This section discusses the main conclusions derived from the research made in this project. With
the results of the data collected from the listening test, conclusions related to headphones sound
exposure and hearing assessment are presented next.
52
6.5. CHAPTER CONCLUSION
The number of subjects tested was 20, which is not considered a number high enough to conclude definitive assumptions regarding either the headphone sound exposure or the hearing assessment.
6.5.1 Conlusions of Headphone Sound Exposure
It was observed that the subjects of both populations adjusted the listening level depending on
the environment. In general terms, a clear progression to higher volume settings is seen when
increasing the external noise to which they are exposed to.
A significant difference was found in the listening level between a very quiet and a noisy environment. However, this difference was not between a very quiet and a moderate, or between a
moderate and a noisy environment.
Nevertheless, it has to be noticed that not all the subjects responds in the same manner. Control
population presents volume settings significantly higher than the target population for each environment. This difference can be appreciated specially in a very quiet environment, where the
control population differs 10 dB from the target population.
Related to exposure levels, it can be concluded that the target population is sightly more exposed
than the control. A difference in the exposure time is not clearly observed. In fact, only one
subject, which is YSN, showed a exposure time representative higher compare to the control
population. Therefore, the listening level is the parameter which makes the difference between
both populations in terms of exposure level.
6.5.2 Validity and Conlusions of Hearing Assessment
Regarding the results from the listening test it can be said that the resolution of the methods used
for the hearing assessment implies not very precise results due mainly to two reasons: First, the
number of subjects tested for the investigation, and second the resolution of the audiometry test
and DPOAEs measurements. Selecting a higher precision, better results could be expected.
The accuracy of the Bracketing method presented in Chapter 2 for the pure tone audiometry is
0.2 dB, but this accuracy does not correspond exactly to the method applied in this project. In
that study [34], the attenuation step size varies during the threshold determination from 5 dB
to 2.5 dB. By contrast, in this project a fixed step size of 5 dB is selected. It leads to a non so
good accuracy. The effects of this step size can be reflected in the standard deviation of the
audiometry data. Nevertheless, the right ear of the control population presents at 6 KHz higher
standard deviation compare to the rest of frequencies of that ear. Therefore the performance of
the subjects of that population at 6 KHz is not as homogeneous as for the rest of frequencies.
53
CHAPTER 6. LISTENING TEST RESULTS
By the analysis of the audiometries it is seen that a mild hearing damage is appreciated at 6 KHz
for both populations. However it can not be concluded that it is due to the use of MPs with
headphones because it is present in both populations. For the rest of frequencies, there is not
any high variation respect to the reference 0 dB HL.
It is seems that the hearing threshold of the different populations tend to follow similar frequency patterns giving representative results. By contrast, a DPOAEs mean of a population can
not be representative due to the big differences between the DPOAE levels of subjects within
a same population. This fact can be appreciated in the standard deviation values. Moreover,
some investigations [8] state that a DPOAE is considered present when it is at least 6 dB above
the noise floor. If this criteria is applied to the DPOAEs mean of the populations, DPOAEs are
present. However, this should be applied to each individual instead to an average population.
Therefore observations of the average across the DPOAE values of the ten subjects of each population does not give a clear picture of the effects of the exposure. To have representative results
a large number of subject should be tested.
From the statistical analysis of DPOAE, its is concluded that there is a significant difference
between both populations. Moreover the results show that at high and low frequencies the
DPOAEs levels of the subjects are significantly different. This conclusion is expected since the
DPOAEs have different amplitudes in different frequencies. Specially DPOAEs are not reliable
at low frequencies since the noise of the measurement becomes higher at low frequency overlapping the DPOAEs.
54
C HAPTER 7
C ONCLUSIONS
The aim of the project is to investigate the possible hearing damage due to the use of MPs with
headphones when listening to music. In order to achieve this goal, two groups of subjects denoted as control and target populations were classified. This was done in a pilot test, where the
headphone sound exposure level was calculated by means of the exposure time and the listening
level.
A 33% of the subjects tested, were users highly exposed according to 85 dB exposure level stated
in the Danish legislation. This group was defined as the target population. Moreover, a 47% of
the subjects reported a moderate exposure, whereas a 20% were non exposed. From these two
last groups, either the subjects whose exposure was lower than 75 dB, or the subjects who were
not users of MPs with headphones, are named as the control population.
After this, a listening test was performed for the two populations to analyze the headphone
sound exposure in a more accurately way. In this case, an individual study of the exposure level
in different environments is performed. Moreover a hearing assessment is carried out.
The headphone sound exposure of the listening test consist of a questionnaire and a control
volume adjustment test. From this part, the listening habits, the listening level and the exposure time are obtained. This was done in three different environments which are very quiet,
quiet/moderate and noisy. Concerning to the hearing assessment, an pure tone audiometry and
DPOAE measurements were carried out.
All the data collected from target and control population were compared to investigate the possible differences regarding to the parameters measured. The final conclusions obtained are:
• The contribution of headphone sound exposure to a possible hearing loss (NIHL) is as
important as the contribution of occupational noise exposure, because the sound pressure
levels produced by the listening devices tested can reach values considerably high. These
listening devices were able to produce sound pressure levels range from 91 dB to 109 dB.
• The environment around a person affects the preferred volume setting that this person
select in his MP specially when the environment becomes noisy. Moreover, users of MPs
with headphones tent to listen to music at higher volume settings than the non users.
55
CHAPTER 7. CONCLUSIONS
• Potential users of MPs with headphones are more exposed than the non potential users
mainly because they set higher volume settings, although their exposure times are similar.
• A mild hearing loss was found in the subjects at 6 kHz, however it can not be concluded
that this damage is due to the use of MPs with headphones, because it is present for exposed and non exposed populations.
For all the reasons stated in this chapter, it is concluded that this study only represents a hearing
state from a group which could be considered in terms of hearing, as being within an average
population with a slight hearing loss in the frequency of 6 KHz. Although the subjects, the
exposure times and noise levels measured may not be fully representative of an average study,
there is no reason to believe that the values obtained are atypical for young people.
56
B IBLIOGRAPHY
[1] Cross section of the organ of corti. http://en.wikipedia.org/wiki/Organ_of_Corti .
[2] Sound data base. http://freesound.iua.upf.edu/index .php .
[3] J.L. Flanagan A Glorig, L.H. Whitney and N. Guttman. Hearing studies of noise on telephon operators. Journal of Speech and Hearing Research, 12:169–178, 1969.
[4] M. Baiamonte F. Mauli A. Peretti, F. Pedrielli and A. Farina. Headphone noise: occupational noise exposure assessment for communication personnel. Euronoise Naples 2003,
2006.
[5] Arbejdstilsynet. Atvejledning d.6.1 (sec. 3). http://www.at.dk/sw10715.asp , 2002.
[6] Dave Berriman. Headphones.Audio and hi-fi handboook pp.310-319. Oxford: Newnes,
1998, 3rd edition, 1998.
[7] James Blauert. Spatial Hearing.The Psychophysics of Human Sound Localization. Massachusetts Institute of Technology, revised edition edition, 1997.
[8] A. Laura Clark. Otoacoustic emission testing in the early identification of noise-induced
hearing loss in South African mineworkers. PhD thesis, University of Pretoria, 2004.
[9] W. Clark. Noise exposure from leisure activities: A review. Journal of the Acoustical
Society of America, 90:175–181, September 1991.
[10] P. Olkinuora E. Airo, J. Pekkarinen. Listening to music with earphones: An assessment of
noise exposure. Acustica - Acta Acustica, 82:885–894, February 1996.
[11] B. J. Fligor. Does earphone type affect risk for recreational noise-induced hearing loss?
Ear and Hearing, 25(6):513–527, 2006.
[12] E. Andrade F.P. Marques. Exposure to occupational noise: otoacoustic emissions test
alterations. Rv. Bras. Otorrinolaringology,, 72(3):362–366, 2006.
[13] Stanley A. Gelfand. Essentials of Audiology. Thieme, fist edition, 1997.
[14] Stanley A. Gelfand. Hearing: An Introduction to Psychological and Physiological Acoustics. Dekker, 1998.
[15] Thomas Graven-Nielsen. Personnal communication with Thomas Graven-Nielsen. PhD
thesis, Laboratory or experimental pain research center for sensory-motor interaction, Aalborg University, 2005.
57
BIBLIOGRAPHY
[16] James Wilbur Hall. Handbook of Otoacoustic Emissions. Thomson Delmar Learning,
2000.
[17] Manual head and torso simulator. brüel & kjær. type 4128c.
[18] Toni Hirvonen. Headphone Listening Methods. PhD thesis, Helsinki University of Technology, 2002.
[19] Standard iec 268-7. sound system equipment-headphones and earphones, 1996.
[20] Standard iec 61672-1. electroacoustics - sound level meters - part 1: Specifications, 2002.
[21] Standard iso 11904-1. acoustics - determination of sound immission from sound sources
placed close to the ear. part 1: technique using a microphone in a real ear(mire technique,
2004.
[22] Standard iso 11904-2. acoustics - determination of sound immission from sound sources
placed close to the ear. part 2: Technique using a manikin, 2004.
[23] Standard iso 1999. acoustics - determination of occupational noise exposure and estimation
of noise-induced hearing impairment, 1990.
[24] Standard iso 389-7. acoustics - reference zero for the calibration of audiometric equipment. part 7: Reference threshold of hearing under free-field and diffuse-field listening
conditions, 2005.
[25] Standard iso 389-8. acoustics - reference zero for the calibration of audiometric equipment.
part 8: Reference equivalent threshold sound pressure levels for pure tones and circumaural
headphones, 2004.
[26] Standard iso 8253-1. acoustics - audiometric test methods - part 1: Basic pure tone air and
bone conduction threshold audiometry, 1993.
[27] Standard itu t 1993. international telecommunication union recommendation p.57, 1993.
[28] L. M. Heller J. Lapsley, L. Marshall and L.M. Hughes. Low-level otoacoustic emissions
may predict susceptibility to noise-induced hearing loss. April 2006.
[29] K. Broughton Jacqueline A. Assessment of the noise exposure of call centre operators.
British Occupational Hygiene Society, 46(8):653–661, 2002.
[30] K. Burk K. Yaremchuk, L. Dickson. Noise level analysis of commercially available toys.
Department of Audiology. Henry Ford Hospital, March 1997.
[31] D.T. Kemp. Otoacoustic emissions, their origin in cochlear function, and use. British
Medical Bulletin,, 63:223–241, 2002.
[32] E. L. LePage and N.M. Murray. Latent cochlear damage in personal stereo users: a study
based on click-evoked otoacoustic emissions. The Medical Journal of Australia, 169:599–
592, 1998.
58
BIBLIOGRAPHY
[33] L. M. Luxon. Disorders of hearing and balance. Reviews in Clinical Gerontology; University College London, UK, 8:31Ű43, 1998.
[34] Morten Lydof. The threshold of hearing & contours of equal loudness - a study of measuring methods and normal hearing. PhD thesis, Aalborg University, 1999.
[35] V. V. Mattila and N. Zacharov. Gls Ű a generalised listener selection procedure. Journal
of the Audio Engineering Society, 49(546), 2001.
[36] Douglas C. Montgomery. Design and Analysis of Experiments. John Wiley & Sons Inc,
6th edition, 2005.
[37] Brian C. J. Moore. An Introduction to Psycology of Hearing. Elsevier Academic Press,
5th edition, 2004.
[38] Rodrigo Ordo nez. Temporary Changes in Human Hearing Caused by Intenses Sounds.
PhD thesis, Aalborg University, 2005.
[39] D. Nondahl. Recreational firearm use and hearing loss. American Medical Association,
March 2007.
[40] C.A. Poldy. Headphones.Loudspeaker and Headphone Handbook pp. 493-574. Focal
Press, 2nd edition, 1994.
[41] D. Hammershøi R. Ordoñez, K. Reuter. Hearing damage by personal stereo: A literature
review. Acoustics Department, Aalborg University, 2006.
[42] D. Hammershøi R. Ordoñez, K. Reuter. Sound exposure by personal stereo, field study of
young people in denmark. Acoustics Department, Aalborg University, 2006.
[43] Karen Reuter. Over-Exposure Effects on the Distortion Product OtoAcoustic Emission:
Broadband and Finestructure. PhD thesis, Aalborg University, 2006.
[44] H.S. Cohen R.W. Alexander, A.H. Koening and C.P. Lebo. The effects of noise on telephon
operators. Journal of Occupational Medicine, 21(1):21–25, 1979.
[45] L. Wang and N. Le Goff. Investigation of the properties of otoacoustic emissions. PhD
thesis, Aalborg University, 2002.
59
Part II
Appendix
61
A PPENDIX A
A SSESSMENT T ECHNIQUES
PARAMETERS
This appendix shows a summary of the parameters selected for both hearing assessment methods
selected in Chapter 2.
A.1 Audiometry
• Method: Fixed discrete pure tone audiometry where, for each frequency, the sound is
presented monaurally through headphones to the subject.
• Sound stimuli: Pure tones of a time length of 300 ms are presented at 250,750, 500, 1000,
1500, 2000, 3000, 4000, 6000 and 8000 Hz.
• Patient instructions: The subject is instructed to press the button when hearing the sound
stimulus, and to not press it when do not hear it.
• Familiarization process: A familiarization is carried out to let the subject get used to the
task. This is done by means of a threshold determination using a pure tone of 40 dB. After
the familiarization, the first tone presented is played at higher level than the level recorded
when the button was not pressed in the last threshold determination.
• Attenuation step size: The sound stimulus is presented at an audible level of 40 dB. Then,
if the subject is able to hear the stimulus, a descent by steps of 10 dB is performed until
the subject does not hear it. Then an increasing of 5 dB is performed.
• Threshold determination: The threshold is determined when the subject responds twice in
a row the same level before he does not hear the signal.
A.2 DPOAE Measurements
• OAEs: DPOAEs are selected, where the distortion product at 2 f 1 − f 2 due to the intermodulation between two tones along the basilar membrane are measured.
• A fixed frequency ratio of f 2 / f 1 = 1.22 is set for all the measurements.
63
APPENDIX A. ASSESSMENT TECHNIQUES PARAMETERS
• The level of the two pure tones was 65 and 45 dB SPL for L1 and L2 respectively. This
values are kept constant for all the measurements.
• The equipment is able to test frequencies from 635 to 3943 . The frequencies are obtained
from f 2 = 1001-6165 Hz. This allows to test the state of the hearing at 635, 696, 769, 830,
903, 1001, 1074, 1172, 1270, 1404, 1526, 1648, 1807, 1965, 2148, 2344, 2563, 2795,
3027, 3308, 3625, and 3493 kHz.
64
A PPENDIX B
T YPES OF HEADPHONES
This appendix present a classification of the different types of headphones according to the
standard ITU-T-1993 [27].
B.1 Introdution
Headphones (also known as earphones, stereo phones or headsets) are a pair of transducers that
receive an electrical signal from a MP or receiver and use speakers placed in close proximity to
the ears to convert the signal into audible sound waves [19].
Important advantages of these devices are their privacy and portability characteristics. Moreover
headphones do not excite room resonances, therefore it gives a more accurate acoustic sense.
Transducers commonly used in headphones are isodynamic, dynamic for example moving-coil,
electrostatic and electret. A more deep explanation of the mechanical and electrical properties of
headphones and transducers can be found in Poldy [40] but it is not examined in this appendix.
B.2 Headphone Categories
Headphones can be divided in five categories based on their structural properties as it is explained in the standard ITU-T-1993 [27].
Likewise, headphones can be classified as open or closed headphones.
Open headphone refers to an intentional leakage built in the back of the headphone cup. Therefore this kind of headphones are provided of an intentionally acoustic path between the external
environment and the ear canal [27]. By contrast, closed earphones are characterized because
they prevent any acoustic coupling between the external environment and the ear canal [27].
Although they are not completely airtight they can be considered nominally sealed. This kind of
headphones are very good at containing the sound away from others and insulating the listener
from outside sounds.
65
APPENDIX B. TYPES OF HEADPHONES
Standard ITU-T-93 [27] states that circum-aural, supra-aural, intra-concha and insert headphones can be open and closed, whereas supra-concha headphones can only be open.
Circum-aural Headphones
Circum-aural headphones are defined as those headphones which enclose the pinna and seat on
the surrounding surface of the head. Contact to the head is normally maintained by compliant
cushions. This kind of headphones may touch, but not significantly compress the pinna [27].
So long as the seal is effective, the frequency response is essentially flat down to a low frequency
which is dependent only on the degree of sealing. A poor seal due to inadequate headband pressure can cause modifications in the headphone response.
The main disadvantage of this type of headphones is that they tent to be heavier and required
greater headband pressure. Moreover they also can make the ears hot and uncomfortable [6]
Supra-aural Headphones
Supra-aural headphones are defined according to the standard ITU-T-1993 [27] as headphones
which rest upon the pinna and have an external diameter or maximum dimension of at least
45 mm.
Compare to circum-aural headphones, supra-aural models are lighter, with smaller cushions, and
do not surround the whole ear. They only cover the concha. The cushion is flat and it is placed
on top of the ear. Therefore supra-aural headphones are more comfortable than circum-aural
headphones. However the frequency response of a supra-aural headphone is more dependent on
the headphone placement than in the case of circum-aural headphones [40].
Intra-concha headphones
Intra-concha headphones are intended to rest within the concha cavity of the ear. They have an
external diameter (or maximum dimension) of less than 25 mm but are not made to enter the ear
canal [27].
This kind of headphones are inserted at the entrance of the ear canal and are supported by the
cartilage of the concha. Compared to the previous types this model is very smaller and portable.
Nevertheless they can be uncomfortable to some people and due to the size.
Insert headphones
Insert headphones are those headphones designed to enter partially or completely into the ear
canal [27], thus the sound is almost conducted directly into the ear canal. These models provide
a good insulation from external sound. They are characterized because the small size of their
transducers. This type of headphones are usually needed for professional issues.
66
A PPENDIX C
E XPOSURE L EVEL C ALCULATION
This appendix shows how to calculate the LEX,8h from the headphones of a MP using a manikin
(head and torso simulator). Standards ISO 11904-2 [22] is followed to calculate the LFF,Aeq
produced by the headphones. Moreover the standard ISO-1999 [23] is also applied to calculate
the LEX,8h . All these values are in dB SPL.
C.1 Method
The calculation of the LEX,8h can be divided in the next three steps:
• Measurement of Leq,Te in Third Octave Bands:
For each of the ear simulators integrated in the manikin, the Leq,Te is measured in onethird octave frequency bands when the manikin is exposed to the sound played through
the headphones.
• Free Field Correction:
Each of the band levels measured before with the manikin is corrected to convert the Leq,Te
inside of the ear, to LF F,eq . In order to do this correction the values show in Figure C.5 are
subtracted from the Leq,Te as it shows in the Equation C.1.
LF F,eq = Leq,Te − ∆LF F
(C.1)
• A-weighted Correction:
After the free field correction, an A-weighted correction is performed to obtain the equivalent continuous A-weighted sound pressure level. It is denoted as LF F,Aeq . Equation C.2
shows the expression of the correction and Figure C.6 shows the values that are added.
LFF,Aeq = LFF,eq + A f
(C.2)
• Calculation of LFF,Aeq :
Since the LF F,Aeq is obtained in third octave bands, a conversion to a value which does
not depend on the frequency is needed. This is done by means of the expression shown in
Equation C.3.
67
APPENDIX C. EXPOSURE LEVEL CALCULATION
LFF,Aeq = 10 log ∑ 10LF F,Aeq
(C.3)
f
• Calculation of LEX,8h :
In order to calculate the sound exposure LEX,8h , Equation C.4 has to be applied
Te
LEX,8h = LFF,Aeq + 10 log
To
(C.4)
where Te is the time that the person is exposed to his headphones in hours per day and To
corresponds to 8 hours.
As an example, the effect of these different steps has been applied to a measurement carried out
with a portable MP and its headphones. Next sections show the equipment, the setup and the
results.
C.2 Setup and Equipment
Table C.1 lists the equipment used in these measurements and Figure C.1 illustrates the setup.
The measurements are done using the control volume position 19 of the portable MP, with the
music sample “Help” from “The Beetles” and with an integration time for Leq,Te measurement
of 60 seconds. Furthermore a exposure time of 4 hours per day is used to calculate the LEX,8h .
C.3 Results
Since the data of Leq,Te, LFF,eq , LF F,Aeq are frequency-dependent, the results are shown in third
octave bands. Moreover an average between the measurements obtained from right and left ear
is done for each plot.
Figure C.2 shows the Leq,Te measured by the microphones placed on the ears of the manikin.
Figure C.3 illustrates the LF F,eq after the free field correction. Figure C.4 shows the LFF,Aeq after
the A-weighted correction.
Finally, the calculated value of the LF F,Aeq is 103.16 dB SPL, and the LEX,8h is 100.15 dB SPL.
68
C.3. RESULTS
Item
Head and Torso Simulator
Portable MP
Headphones
Measurement System
PC
Model
B&K 4128
Creative MuVo V200
Creative MuVo V200
0.1 dB Harmonie
Fujitsu (laptop) -1600MB RAM, Intel M 760MHz
Table C.1: Equipment used for the measurements of the
Leq,Te
−
out
in
out
in
MP
out
PC
01 dB
Symphonie
Head and Torso Simulator
B&K 4128
Figure C.1: Setup used for the measurements of the
69
Leq,Te
LAB-Nr:
08453060
215741
215741
33964
33964
APPENDIX C. EXPOSURE LEVEL CALCULATION
120
100
Leq,60s [dB SPL]
80
60
40
20
0
Figure C.2:
Leq,60s
16
31.5
63
125
250
500
1k
Frequency [Hz]
2k
4k
8k
16k
in third otave bands measured by the mirophones plaed on the ears of the manikin
with the Creative MuVo V200 portable MP and its headphones. The ontrol volume position
seleted for the measurement is 19 and the integration time is
graph is an average of the right and left ear.
70
60 seconds.
The data show in this
C.3. RESULTS
120
100
LeqFF,60s [dB SPL]
80
60
40
20
0
Figure C.3:
LeqFF,60s
16
31.5
63
125
250
500
1k
Frequency [Hz]
2k
4k
8k
16k
in third otave bands alulated after the free eld orretion. The data after the free
eld orretion is shown in Figure C.2. The data show in this graph is an average of the right
and left ear.
120
100
LAeqFF,60s [dB SPL]
80
60
40
20
0
Figure C.4:
LAeqFF,60s
16
31.5
63
125
250
500
1k
Frequency [Hz]
2k
4k
8k
16k
in third otave bands alulated after the free eld orretion and the A-weighted
orretion. The data after the A-weighted orretion is shown in Figure C.3. The data show in
this graph is an average of the right and left ear.
71
APPENDIX C. EXPOSURE LEVEL CALCULATION
16
14
12
∆ [dB SPL]
10
8
6
4
2
0
−2
1
10
2
10
3
10
Frequency [Hz]
4
10
5
10
Figure C.5: Standardized values of the free eld orretion for ear measurements of the manikin. The values
that are subtrated are those whih orrespond with the third otave bands frequenies.
10
A−weighted corrections [dB SPL]
0
−10
−20
−30
−40
−50
−60
1
10
2
10
3
10
Frequency [Hz]
4
10
5
10
Figure C.6: Standardized values of the A-weighted orretion for ear measurements of the manikin.
The
values that are subtrated are those whih orrespond with the third otave bands frequenies.
These values are obtained from the standard IEC 61672-1 [20℄.
72
A PPENDIX D
P ILOT T EST M EASUREMENTS
This appendix describes the measurements needed for the pilot test described in Chapter 4. The
normalization of the music samples and the measurements for the selection of the portable MP
and the two headphones used in the pilot test are explained. Finally, the LFF,Aeq produced by the
selected MP with both headphones when the portable MP is set to different volume settings is
carried out.
D.1 Method
The measurements are divided in three different parts:
• Measurements related to the normalization of the music samples selected for the pilot test.
• Measurements in order to select a portable MP and a pair of headphones.
• Measurements of the LFF,Aeq produced at the different volume control positions of the
portable MP and the headphones selected.
D.2 Setup and Equipment
The measurements are carried out in the listening room A at Aalborg University. Figure D.1
illustrates the setup and Table D.1 lists the equipment. The same setup and equipment is used
for all the measurements described in this appendix.
The measurements are performed using a head and torso simulator. This device is equipped
with a set of soft artificial pinnas, an artificial ear canal and an occluded artificial ear which
allows to simulate the sound pressure level in the inner part of a real human ear. The pinnas
have a hardness very close to that of a real human ear, therefore the head and torso simulator
enables easy and realistic mounting of headphones [17]. In addition headphones evaluation can
be performed because the artificial ear provides the correct acoustic loading [17].
The portable MP is connected to an adapter that splits the signal in two. One of its outputs is
connected to the headphones and other is connected to a voltmeter for measuring the output
voltage of the MP. The headphones are placed on the head and torso simulator, thus the microphone of the artificial ear measures the sound pressure level. The measurements are recorded in
73
APPENDIX D. PILOT TEST MEASUREMENTS
the time domain with the 0.1 dB Harmonie Measurement System and later they are proccessed
in MATLAB in order to obtain the LAeq,60s. The integration time is set to 60 seconds which is
the time duration of all the music samples.
The head and torso simulator has two artificial ear inserted in the right and left side of the head,
therefore two measurements were obtained in each recording.
Item
Head and Torso Simulator
Model
B&K 4128
Creative MuVo V200
Supratech JazzFree
Creative
Super PowerBass CBX-15
Supratech JazzFree
Miniheadphones a11P1
Sony MDR
0.1 dB Harmonie
Fujitsu (laptop) -1600MB RAM, Intel M 760MHz
-
PMP
Headphones
Measurement System
PC
Voltmeter
Table D.1: Equipment used for the pilot test measurements.
Voltimeter
+
Adapter
−
in
out
MP
out
in
Adapter
out
01 dB
Symphonie
Head and Torso Simulator
B&K 4128
Figure D.1: Setup used for the pilot test measurements.
74
PC
LAB-Nr:
08453060
215741
215741
2741
33964
33964
-
D.3. RESULTS
D.3 Results
D.3.1 Normalization of the Musi Samples
The different selected music samples that are presented to the subject in the pilot test were
collected from different information sources. All the music samples are single-channel WAVE
files, with a sample frequency of 44.1 kHz, 16 bits per sample and 60 seconds of duration. The
problem with these stimuli was the deficiency to control the level of the recordings beforehand.
Due to this, when a music sample is played in the portable MP with the same volume control
position, the equivalent level of the sound produced at the headphones is different depending on
the level used when the recordings of the songs were made. Therefore a normalization of the
music samples used in the pilot test is needed. For example Figure D.2 shows 30 seconds of two
different samples before the normalization. It is observed that the energy content of these two
signals is different.
1
Amplitude
0.5
0
−0.5
−1
0
5
10
15
Time(s)
20
25
30
0
5
10
15
Time(s)
20
25
30
1
Amplitude
0.5
0
−0.5
−1
Figure D.2: The upper gure orresponds to rst 30 seonds of the musi sample Help from The Beetles"
and the lower gure orresponds to the rst 30 seonds of the musi sample Hips don't lie
from Shakira. Both gures show the time signals before the normalization.
The normalization is applied to all the music samples to get the same LAeq,60s. This is done
following the definition of LAeq,T stated in the standard ISO 1999 [23]. Equation D.1 shows the
expression where T is the duration of the recording, PA (t) is the instantaneous sound pressure
(A-weighted) at time t in Pascals, and P0 is the standard reference pressure which is 20µPa. The
A-weighted is performed in order to get the signal as perceived by the ear.
LAeq,T
Z T 2
PA (t)
1
= 10 log
dt
T 0 P02
(D.1)
The procedure to normalize all the music samples is based on the calculation of the LAeq,60s of
one of the samples, and then adjust the rest to have the same LAeq,T than this first sample, which
75
APPENDIX D. PILOT TEST MEASUREMENTS
is the reference. This is implemented in MATLAB.
From Equation D.1, an expression for the normalization can be derived by means of a level
difference dBdi f f (in dB) between two equivalent continuous A-weighted sound pressure levels,
LAeq1,T and LAeq2,T . This is show in Equation D.2 where P2 (t) is the time function (instantaneous
pressure values) of the signal to be normalized and P1 (t) is the time function of the reference
signal. Both, P1 (t) and P2 (t) are in Pascals.
dBdi f f = LAeq1,T − LAeq2,T = 20 log
Z T
P1 (t)
0
(D.2)
P2 (t)
Thus, the equation that describes the normalization can be obtained solving P2 (t) from Equation
D.2. The solution is described in Equation D.3.
P2 (t) = P1 (t).10−
dBdi f f
20
(D.3)
The reference signal chosen is the music sample “Help” from the “The Beetles” which has a
LAeq,60 of 96.03 dB. Therefore, by applying Equation D.3 to all the music samples, they all are
adjusted to this equivalent continuous A-weighted sound pressure level.
In order to test the normalization, the LAeq,60s of the stimuli is calculated in MATLAB after and
before the normalization. Figure D.3 shows 30 seconds of the two stimuli shown in Figure D.2,
but after the normalization.
1
Amplitude
0.5
0
−0.5
−1
0
5
10
15
Time(s)
20
25
30
0
5
10
15
Time(s)
20
25
30
1
Amplitude
0.5
0
−0.5
−1
Figure D.3: The upper gure orresponds to rst 30 seonds of the sound stimulus Help from The Beatles
and the lower gure orresponds to the rst 30 seonds of the sound stimulus Hips don't lie
from Shakira. Both gures show the time signals after the normalization. The
ases is
96.03 dB
76
LAeq,60s
in both
D.3. RESULTS
D.3.2 Measurements for the Seletion of the MP and the Headphones
In order to fairly select a portable MP and a pair of headphones, all possible combinations between the two portable MP and the five sets of headphones shown in Table D.1 are tested. Each
combination is identified by a reference name as it can be seen in Table D.2. For instance LD3
would correspond to the Listening device (LD) formed by the Creative MuVo V200 portable
MP and the Super PowerBass CBX-15 headphones.
A requirement for the portable MPs that are tested is that they must have a digital volume setting
for a easier control of the volume set by each subject.
Regarding the headphones, they must be portable, comfortable and easy to place. The types of
headphones selected to be tested are supra-concha and intra-concha earphones. It is decided not
to use circum-aural headphones because they are not considered portable. In addition, insert
earphones are neither used because they are not comfortable enough furthermore, these devices
are associated with professional issues [18].
Item
LD1
LD2
LD3
LD4
LD5
LD6
LD7
LD8
LD9
LD10
PMP
Creative MuVo V200
Creative MuVo V200
Creative MuVo V200
Creative MuVo V200
Creative MuVo V200
Supratech JazzFree
Supratech JazzFree
Supratech JazzFree
Supratech JazzFree
Supratech JazzFree
Types of Headphone
Creative
Supratech JazzFree
Super PowerBass CBX-15
Miniheadphones a11P1
Sony MDR
Creative
Supratech JazzFree
Super PowerBass CBX-15
Miniheadphones a11P1
Sony MDR
Table D.2: Possible ombinations between the two MP and the four headphones with their orresponding
assigned name. The measurements are done aording to these ombinations for the pilot test
measurements.
Two LDs are selected according to the next requirements.
• First, the selected LDs must produce the highest sound pressure level among all the possible LDs.
• Second, the two LDs chosen must have different types of headphones.
The selection is made according to the results of the next measurements.
• Sensitivity of the headphones.
• Maximum LFF,Aeq delivered by the LDs.
77
APPENDIX D. PILOT TEST MEASUREMENTS
Sensitivity of the Headphones:
The sensitivity of the headphones is measured playing in the portable MP a pure tone signal at
1000 Hz. Then, the sound pressure level is directly measured by means of the software of the
Symphonie system and the voltage is measured in the voltmeter. From these two measurements
a sensitivity value can be derived. This is done for all the headphones and for both cups of
each headphone. Since it is a headphone measurement, either the Supratech JazzFree MP or the
Creative MuVo V200 MP can be used as a sound source because it does not change the results.
In Table D.3 the values of the sensitivities measured for the different headphones are shown.
Headphones Type
Creative
Supratech JazzFree
Super PowerBass CBX-15
Miniheadphones a11P1
Sony MDR
Voltage
(Vrms )
0.268
0.271
0.246
0.270
0.251
SPL
right cup (dB)
101.7
87.0
90.7
76.7
101.6
SPL
left cup (dB)
102.5
86.5
91.5
81.3
103.0
Sensitivity
right cup (dB/V)
113.93
97.84
103.68
92.67
115.00
Sensitivity
left cup (dB/V)
113.13
98.34
102.88
88.07
113.60
Table D.3: Measured sensitivity of the dierent headphones. The measurements are performed for both ups
playing a pure tone at
1000 Hz
using the Suprateh JazzFree MP. The sound pressure level is
measured only at that frequeny.
Maximum LFF,Aeq delivered by the LDs:
The music sample used is the song "The Beetles" which is one of the stimuli signals presented
to the subject as it is explained in the design of the pilot test in Chapter 4. However any of
the music samples could be used because they are all normalized. The integration time for the
calculation of LF F,Aeq is set to the duration of the music sample which is 60 seconds.
Listening
Device
LD1
LD2
LD3
LD4
LD5
LD6
LD7
LD8
LD9
LD10
Table D.4: Maximum
LFF,Aeq
LFF,Aeq (dB)
left cup
109.77
104.57
105.31
97.29
105.80
109.43
103.47
102.43
96.47
104.60
LFF,Aeq (dB)
right cup
107.60
104.92
101.96
90.78
104.38
106.24
98.25
97.45
91.90
103.73
when the volume is set to the maximum for the dierent LDs.
sample used is the song titled "Help" from "The Beetles".
78
The musi
D.3. RESULTS
Final Selection
From the results showed in table D.4 it can be concluded that LD1, which gives 109.77 dB and
107.60 dB for the right and left cup respectively, are the devices that produce the highest LFF,Aeq .
Taking in account that the other LD selected must have different type of headphones, the next
LD that gives a high LF F,Aeq is LD5. It produces 105.80 dB and 101.96 dB for right and left
cup respectively. In addition, the headphones of LD1 and LD5 are the ones with the highest
sensitivity, therefore they provide the highest sound pressure level among all the headphones. It
can be appreciated in table D.3.
In conclusion, even though no big different among LD is seen, LD1 and LD2 are selected according these measurements.
D.3.3 Volume Setting Measurements
With the two LDs selected in the last section, a measurement of the LFF,Aeq for different control
volume positions is done. The measurements are performed moving the digital control position
from the minimum to the maximum volume. The signal used is again the musical excerpt from
"The Beetles" and titled "Help" and the integration time for the calculation of LFF,Aeq is the
duration of the music sample (60 seconds).
The data obtained in the volume setting measurements is shown in Tables D.5 and D.6 for the
two LDs. Moreover Figure D.4 shows this data in a graph.
79
APPENDIX D. PILOT TEST MEASUREMENTS
Control Volume
Position
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
Table D.5: Maximum
LFF,Aeq (dB)
left cup
75.54
75.37
76.83
77.57
79.09
80.38
82.06
83.71
85.38
87.16
89.12
91.04
92.97
94.98
96.99
98.86
100.68
102.35
103.91
105.34
106.55
107.66
108.64
109.07
109.77
LFF,Aeq
LFF,Aeq (dB)
right cup
73.30
73.75
75.34
76.19
77.83
79.19
80.92
82.64
84.40
86.19
88.18
90.12
92.02
94.03
96.00
97.78
99.45
101.00
102.42
103.71
104.78
105.75
106.57
106.96
107.60
Mean LF F,Aeq (dB)
74.42
74.56
76.08
76.88
78.46
79.79
81.49
83.17
84.9
86.68
88.65
90.58
92.50
94.50
96.50
98.32
100.06
101.67
103.17
104.52
105.66
106.71
107.60
108.02
108.68
when the volume is set to dierent positions for LD1.
80
D.3. RESULTS
Control Volume
Position
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
Table D.6: Maximum
LFF,Aeq (dB)
left cup
70.64
71.03
71.96
72.44
74.43
74.32
75.44
76.67
78.18
79.82
81.58
83.42
85.37
87.25
89.24
91.19
93.21
95.16
97.14
99.03
100.80
102.46
104.06
104.81
105.96
LFF,Aeq
LFF,Aeq (dB)
right cup
66.72
67.24
68.71
69.06
71.19
71.27
72.52
73.84
75.42
77.14
79.02
80.82
82.86
84.71
86.68
88.62
90.66
92.61
94.58
96.41
98.15
99.71
101.25
101.97
105.96
Mean LF F,Aeq (dB)
68.68
69.14
70.34
70.75
72.81
72.80
73.98
75.26
76.80
78.48
80.30
82.12
84.12
85.98
87.96
89.91
91.93
93.89
95.86
97.72
99.48
101.08
102.65
103.39
105.96
when the volume is set to dierent positions for LD5.
81
APPENDIX D. PILOT TEST MEASUREMENTS
120
Listening Device1
Listening Device5
110
Leq FF (dBA)
100
90
80
70
60
Figure D.4: Mean
0
1
LFF,Aeq
4
7
10
13
16
Control Volume Positions
19
22
25
produed by the LD1 and the LD5 measured at the headphones using an artiial
head and torso simulator. Cirles represents the measured values and the urve is alulated by
interpolation. An average between the measurements of the right and left up is done. This
data an be also seen in Tables D.5 and D.6 respetively
82
A PPENDIX E
P ILOT T EST I NSTRUCTIONS
Welcome to our Listening Test. The pilot test you are about to take part in is carried out for
the final master thesis project performed by group 1066 of the Acoustic department of AAU. It
studies the exposure level and the listening habits when music players with headphones are used.
Task 1: In this task you have to:
• Select one of two different headphones showed.
• Select one of seven music samples presented.
• Adjust the control volume setting of a given portable music player using the headphones
and the music sample elected before.
Two different types of headphones are going to be presented. You have to choose the one that
looks more similar to the headphone that you use normally. In case that you are not a music
player user, you must select that set of headphones that you would use if you were going to
listen to music with a music player.
In next table there is a list of music samples of varied styles. You have to choose the sample that
you would like to listen to.
Author
The Beetles
Milk Inc.
Fugees
Woody Allen
Shakira
U2
Vivaldi
Title
Help
In my eyes
Ready or not
Come On and Stomp
Hips dont’t lie
Sunday Bloody Sunday
The Four Seasons
Table E.1: Musi samples
Task 2: In this task you have to fill out a questionnaire.
All your data will be stored and treated confidentially.
Thanks for your participation
Acoustics Department. Group 1066
83
A PPENDIX F
P ILOT T EST Q UESTIONNAIRE
This is a questionnaire carried out for the final master thesis project performed by group 1066
of the Acoustic department of Aalborg University. If you want to participate in this listening
experiment, please fill in the form bellow. Your data will be stored and treated confidentially.
F.0.4 Personal data:
• Name and Surname:....................................................................................................
• Age:....................
• Genre: Woman
Man
F.0.5 Prior experiene:
1 Do you have a known history of hearing damage diagnosed by a medical doctor?
Yes
No
I dont’t know
If last answer was Yes, which and when?............................................................................
...............................................................................................................................................
2 Have you ever had an ear operation?
Yes
No
I dont’t know
If last answer was Yes, which and when?...............................................................................
..............................................................................................................................................
3 Do you ever had any hearing problem i.e. infections, ear noises, drainage...?
Yes
No
I dont’t know
If last answer was Yes, how often?.......................................................................................
..............................................................................................................................................
4 Are you suffering a cold?
Yes
85
APPENDIX F. PILOT TEST QUESTIONNAIRE
No
I dont’t know
5 Have you ever had your hearing tested before?
Yes
No
I dont’t know
If last answer was Yes, when, where and how did it go?.........................................................
..............................................................................................................................................
6 Does/Did anyone in your immediate family have hearing disorder?
Yes
No
I dont’t know
If last answer was Yes, please explain................................................................................
..............................................................................................................................................
7 Have you ever taken any medicine or another type of drugs that might have influence on
your hearing?
Yes
No
I dont’t know
If last answer was Yes, which and when?...........................................................................
..............................................................................................................................................
8 Have you ever been exposed to loud sounds i.e. explosions, fireworks, shootings, firearms...?
Yes
No
I dont’t know
9 If last question was Yes:
– Please explain if you felt any pain in your ears or sudden lost of hearing?.........................
..........................................................................................................................................
– Which kind of sound and how often have you been exposed?..........................................
.........................................................................................................................................
– Did you use hearing protectors?
Yes
No
I dont’t know
If last answer was Yes, for how long?............................................................................
..........................................................................................................................................
10 Have you ever had a job where you need to use headphones i.e. call center, dj...?
Yes
86
No
I dont’t know
If last answer was Yes, for how long?...................................................................................
................................................................................................................................................
11 Have you ever worked in a very noisy environment where you could not talk easily with
other people?
Yes
No
I dont’t know
12 If last question was Yes:
– Which kind of noise and in which environment?.............................................................
.........................................................................................................................................
– How long and how often?.................................................................................................
.........................................................................................................................................
– Did you use hearing protectors?
Yes
No
I dont’t know
If last answer was Yes, for how long?..........................................................................
.......................................................................................................................................
F.0.6 Listening habits:
13 Do you listen to music using music players with headphones?
Yes
No
If last answer was No, please go to question number 23
14 If the music player with headphones that you use is a portable music player i.e. mp3,
mobile telephone, portable radio, PDA ... please, fill out next table choosing how often
and for how long time you have been using it:
How often?
weekly
monthly
Day
1
2
3
4
5
6
7
Time in hours per day
1 or less
1-2
2-4
4-6
6-8
8-10
more than 10
87
How long?
1 month
1-6 months
6-12 months
1-2 years
2-5 years
5-10 years
more than 10 years
APPENDIX F. PILOT TEST QUESTIONNAIRE
15 If the music player with headphones that you use is a personal computer or a laptop please
fill out next table choosing how often and for how long time you have been using it:
How often?
weekly
monthly
Day
1
2
3
4
5
6
7
Time in hours per day
1 or less
1-2
2-4
4-6
6-8
8-10
more than 10
How long?
1 month
1-6 months
6-12 months
1-2 years
2-5 years
5-10 years
more than 10 years
16 If the music player with headphones that you use is a TV or a DVD, please fill out next
table choosing how often and for how long time you have been using it:
How often?
weekly
monthly
Day
1
2
3
4
5
6
7
Time in hours per day
1 or less
1-2
2-4
4-6
6-8
8-10
more than 10
How long?
1 month
1-6 months
6-12 months
1-2 years
2-5 years
5-10 years
more than 10 years
17 If the music player with headphones that you use is a Hi-Fi equipment or similar, please
fill out next table choosing how often and for how long time you have been using it:
How often?
weekly
monthly
Day
1
2
3
4
5
6
7
Time in hours per day
1 or less
1-2
2-4
4-6
6-8
8-10
more than 10
How long?
1 month
1-6 months
6-12 months
1-2 years
2-5 years
5-10 years
more than 10 years
18 If the music player with headphones that you use is a mixer table or similar, please fill
next table choosing how often and for how long time have you been using it:
88
How often?
weekly
monthly
Day
1
2
3
4
5
6
7
Time in hours per day
1 or less
1-2
2-4
4-6
6-8
8-10
more than 10
How long?
1 month
1-6 months
6-12 months
1-2 years
2-5 years
5-10 years
more than 10 years
19 Next table contains some places where music players with headphones can be used.
Please fill the table marking how often you use to listen to your music player with headphones and the type of environment which corresponds to the place selected. You only
have to mark those places where you use your music player.
bus
train
bike
motor-bike
car
street
home
university
at work
another place:...........................
1 Rarely
how often?
2 Often 3 Very often
which environment?
1 Quiet 2 Moderate 3 Noisy
20 Do you use to place the two earphones in your ears when you are listening to music with
your music player?
Yes
No
21 If last question was No, which ear do you use to cover?
right
left
22 Which kind of headphones do you use?
Inside the ear
On the ear
Around the ear
23 Are you a musician?
Yes
89
APPENDIX F. PILOT TEST QUESTIONNAIRE
No
If last question was No, please go to question number 26
If last answer was Yes, which kind of instrument do you play?...............................................
24 Are you playing in a:
Rock band
Orchestra
Jazz band
Big band
At home
studio
at work
Another one: .......................................
pub
25 Please, fill next table choosing how often do you play and for how long time have you
been playing
How often?
weekly
monthly
Day
1
2
3
4
5
6
7
Time in hours per day
1 or less
1-2
2-4
4-6
6-8
8-10
more than 10
How long?
1 month
1-6 months
6-12 months
1-2 years
2-5 years
5-10 years
more than 10 years
26 Do you go to discotheques or pop/rock concerts?
Yes
No
27 If last question was Yes, please fill next table choosing how often do you go to discotheques or pop/rock concerts and for how long have you been going to this places.
How often?
weekly
monthly
yearly
Day
1
2
3
4
5
6
7
Time in hours per day
1 or less
1-2
2-4
4-6
6-8
8-10
more than 10
90
How long?
1 month
1-6 months
6-12 months
1-2 years
2-5 years
5-10 years
more than 10 years
This questionnaire is a part of a master thesis which study the hearing and the sound exposure
level when music players with headphones are used. Do you want to take part of our listening
test? In this study your hearing will be tested using common and not risky methods as audiometry and otoacoustic emissions (OAE). Furthermore the sound exposure level when you are using
portable music players with headphones will be measured and finally, you will be asked to fill a
short questionnaire.
Do you want to participate?
Yes
No
Telephone number:.............................................................
e-mail address:......................................................................
Thanks for your participation
Acoustics Department. Group 1066
91
A PPENDIX G
P ILOT T EST S UBJECTS
This appendix lists the subjects who performed the pilot test and their results.
The reference number, the sex, the age and the volume control level setting of each of the subjects can be seen in Table G.1. Moreover the LFF,Aeq , the exposure time in hours per day and the
LEX,8h calculated from the data collected in the pilot test are also presented.
It must be noticed that the subjects who present any hearing condition in the pilot test are not
listed in this table.
The data of Table G.1 is sorted from the minimum LEX,8h to the maximum LEX,8h . Figures G.1
and G.2 plots the data shown in this table. Figure G.1 shows all the subjects whereas figure G.2
shows only the subjects with a LEX,8h different from zero.
110
100
90
80
LEX 8h [dB SPL]
70
60
50
40
30
20
10
0
Figure G.1:
LEX,8h
1
5
10
15
20
25
30
35
Reference number of the subjects
40
45
alulated from the data olleted in the pilot test. The
from the minimum to the maximum
LEX,8h .
93
50
LEX,8h
54
of the subjets is sorted
This data is also shown in Table G.1
APPENDIX G. PILOT TEST SUBJECTS
100
95
90
LEX 8h [dB SPL]
85
80
75
70
65
60
10
Figure G.2:
LEX,8h
15
20
25
30
35
40
Reference number of the subjects
45
50
55
alulated from the data olleted in the pilot test. The subjet with a
are not shown. The
LEX,8h
LEX,8h
of the subjets is sorted from the minimum to the maximum
This data is also shown in Table G.1
94
0 dB
LEX,8h .
of
Reference Number
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
Sex
male
female
male
female
female
female
male
male
female
male
female
female
female
male
male
female
female
male
male
male
female
male
female
male
female
male
male
female
male
female
male
male
male
male
male
male
female
male
male
male
female
male
male
male
female
male
male
male
male
male
male
female
male
female
Age
30
22
27
22
24
24
21
24
31
27
22
31
25
33
25
23
26
34
24
26
21
21
23
21
24
24
22
25
24
21
29
26
28
25
27
24
21
24
25
22
26
22
27
27
24
21
23
26
22
23
22
28
25
22
Control Level Setting
16creative
10sony
13sony
18creative
20sony
12sony
15sony
19creative
17sony
8sony
9creative
1creative
6sony
10creative
12sony
9creative
13creative
13sony
10creative
10creative
10creative
12sony
9creative
16sony
16sony
15sony
13sony
15sony
16sony
10creative
14creative
19sony
13creative
17sony
10creative
16sony
18sony
18sony
12creative
20sony
12creative
21sony
15creative
22sony
19creative
18sony
19creative
20sony
15creative
19creative
21sony
20creative
21creative
22sony
95
Table G.1: List of the subjets who performed the pilot test.
ondition in the pilot test are not listed in this table.
LFF,Aeq
98,32
78,48
84,12
101,67
97,72
82,12
87,96
103,17
91,93
75,26
84,89
74,42
72,80
86,68
82,12
84,89
92,50
84,12
86,68
86,68
86,68
82,12
84,89
89,91
89,91
87,96
84,12
87,96
89,91
86,68
94,50
95,86
92,50
91,93
86,68
89,91
93,89
93,89
90,58
97,72
90,58
99,48
96,50
101,08
103,17
93,89
103,17
97,72
96,50
103,17
99,48
104,52
105,66
101,08
Hours per Day
0
0
0
0
0
0
0
0
0
0
0
0,8
2,13
0,13
0,8
0,6
0,13
1
0,6
0,6
1,2
4
2,6
0,9
0,93
1,6
4
1,7
1,33
3,3
0,6
0,46
1
1,3
5,3
2,6
1,06
1,06
2,5
0,66
3,7
0,5
1,06
0,4
0,26
3,7
0,5
1,8
4,5
1,3
4
1,8
1,8
5,9
LEX,8h
0,00
0,00
0,00
0,00
0,00
0,00
0,00
0,00
0,00
0,00
0,00
64,42
67,05
68,78
72,12
73,64
74,71
75,08
75,43
75,43
78,44
79,11
80,01
80,42
80,56
80,97
81,11
81,23
82,11
82,83
83,26
83,45
83,46
84,04
84,89
85,02
85,11
85,11
85,53
86,88
87,23
87,43
87,72
88,07
88,29
90,54
91,13
91,24
94,00
95,28
96,47
98,05
99,19
99,76
The subjets who presented any hearing
A PPENDIX H
L ISTENING T EST I NSTRUCTIONS
Welcome to our Listening Test!
The listening test you are about to participate is part of a listening experiment that studies the
effects of using music players with headphones on the hearing. You have already participated in
the first phase and now it comes the second phase. In this paper you will find a short description
of the parts of the listening test. Read carefully these instructions and if you have any doubt, do
not hesitate to ask us. Feel free to eat cake and drink coffee or soda while you are reading this
paper:
• FIRST PART:
The first part takes around 5 minutes and consist of a questionnaire and several questions
about your listening habits. It is quite similar to the questionnaire that you filled out in
the first phase of this listening experiment.
• SECOND PART:
This second part takes around 20 minutes and consist of an audiologic evaluation. Your
ears will be tested using two not risky methods which are an audiometric test and otoacoustic emissions measurements. If you are interested on the results of the tests, we can
give them to you in the end of the listening test. After this, you will have a break of 5
minutes. During this time you can leave the room, eat more cake and drink coffee or soda.
• THIRD PART:
This third part will take around 20 minutes. Your task will be to adjust the volume of your
own music player using your own headphones in different environments. If you could not
bring your own devices, we will provided you the music player and the headphones that
you used in the first phase of this experiment.
This is a brief explanation of the listening test. It is very important that you feel relaxed and
comfortable during all the experiment. If for any reason you want to quit the experiment, you
can do it at any time without any explanation.
Thank you very much for your participation.
Group 1066. Acoustic Department
97
APPENDIX H. LISTENING TEST INSTRUCTIONS
First Part: Audiometry Test
• DESCRIPTION:
During this test you will have to wear a pair of headphones. When you place them, be
sure that they fit your ears well. Then different sounds will be presented, first on one ear
and then in the other.
• YOUR TASK:
Your task is to press a button that you will have in your hand whenever you hear something. If no sound is heard, you do not have to press the button. In order to be sure that
you understand this procedure a trial will be performed in the beginning.
• TIPS:
It is very important that you feel relax and concentrated during this test because weak
sounds are going to be presented.
Seond Part: Otoaousti Emissions Measurements
• DESCRIPTION:
In this hearing test, a test probe will be inserted in your ear and some sounds will be
played through it. All the sounds will be played at comfortable levels.
• YOUR TASK:
During this test you do not have to do anything, you just have to be relaxed.
• TIPS:
This measurement is very sensitive to movements of the probe, so you must try to be
quiet, do not move and try not to swallow.
Remember that after this test you will have a short break of 10 minutes when you can enjoy our
cake and coffee.
98
Third Part: Adjustment of the Volume of your Musi Player
using your Headphones in dierent Environments
• DESCRIPTION:
You are going to be conducted to a listening room where a street, a bus and a very quiet
environment are going to be simulated in different moments.
• TASKS:
Your task is to set the volume of your music player using your headphones every time
that an environment is simulated. It is very important that you select the volume that you
would use if you would be in these situations.
You will be sit down in a listening room. The first thing you have to do is to listen to the
environment during 20 or 30 seconds without headphones and with the music player off.
Then, you must place the headphones on, turn on the music and play the song you would
like to listen to. Finally you have to adjust the volume of the music player and let us know
that you finished by means of an intercomunicator.
After the volume setting of each environment, you will have a break of 3 minutes and you
will have to leave the room.This procedure will be repeat 9 times.
It is very important that after each volume setting you do not touch or vary the control
volume of the music player.
The maximum time that you have for the volume settings is 5 minutes per environment.
You can choose any song that you have in your music player , but once you chose it, you
will have to use that song for all the environments.
The first time that you adjust the volume of your music player is taken as trial to make
you get used to the procedure to follow.
• TIPS:
In this part of the listening test you have to trust your feelings and try to imagine that you
are in a bus, a street or a quiet environment. Feel free to set the volume that you would
use, no matter which one it is. Remember that there is not right or wrong settings, so
please try to be honest and show your preferences.
99
A PPENDIX I
I NSTRUCTIONS FOR THE T EST
E XPERIMENTER
This appendix explains the procedure followed by the test experimenter when doing the listening
test with a subject. Basic rules and some general issues that should be well known by the test
experimenter are described in this appendix.
I.1 Basi Rules for the Test Experimenter
It is important that the subject is feeling well and comfortable doing the complete listening test.
In order to do this, different rules that must be followed by the test experimenter are defined in
the next lines:
• The test experimenter must be polite all the time, even if the subject is tired, angry, or
impolite.
• The test experimenter must act as if he is in control of everything during the experiment.
• It is the experimenters task to welcome the subject by means of guiding the subject and
eliminating possible misunderstandings.
• It is the experimenters task to have a good treatment of the subject and explain to him the
needed indications in order to do the listening.
I.2 Test Experimenter Instrutions
The instructions followed by the test experimenter for the different parts of the listening test are
described in this section:
I.2.1 Welome and Introdution
• Welcome the subject and say that we appreciate to have him/her as a subject.
• Conduct the subject to the waiting room.
• Invite the subject to sit down and offer him soda and cake in the waiting room.
101
APPENDIX I. INSTRUCTIONS FOR THE TEST EXPERIMENTER
• Hand the instructions paper to the subject and ask him/her to read it thoroughly.
• Ask if they understand what they read and if they have any questions or objections to it.
• Start the listening test.
I.2.2 First Part of the Listening Test: Personal Questionnaire
• Give the questionnaire to the subject.
• Instruct the subject in the procedure of filling it out.
• Take the questionnaire with you and conduct the subject into the audiometry cabin.
I.2.3 Seond Part of the Listening Test: Bekesy Audiometry and DPOAEs
Measurements
• Instruct the subject where to sit in order to perform the audiometry.
• Give the audiometry instructions to the subject.
• Ask if they understand what they read and if they have any questions or objections to it.
• Be sure that the subject puts on the headphones correctly in terms of left/right.
• Start the familiarization process of the audiometry.
• When finished, ask the subject if everything is all right and if he/she has any questions
regarding the method.
• Perform the audiometry for both ears.
• Save the data in the audiometer with the name of the subject.
• Introduce the subject to the DPOAEs measurements.
• Give the DPOAEs measurements instructions to the subject.
• Ask if they understand what they read and if they have any questions or objections to it.
• Put the plastic globes on the hands and place the test probe in the hear of the subject.
• Be sure that the test probe is correctly placed by testing it in the OAEs measurement
equipment.
• Perform the DPOAEs measurements per each ear and save the data of each subject in the
floppy disc and the hard disc of the computer.
• When the measurement is finished, conduct the subject to the waiting room. Feel free to
small talk but do not discuss any technical test related issues. Wait until after the listening
test is done with discussing the test in general.
102
I.2. TEST EXPERIMENTER INSTRUCTIONS
I.2.4 Third Part of the Listening Test: Control Volume Adjustment
• Introduce the subject to the control volume adjustment part.
• Ask to the subjects about the music player and the headphones they brought. Be sure that
they wrote correctly the model of the devices in the questionnaire as well as the music
sample that they are going to use for this part.
• Give the control volume adjustment instructions to the subject.
• Be sure that the subject understand what he read and ask if he has any questions.
• Conduct the subject to the listening room and show him where to sit.
• Remain to the subject that he has to listen to the environment during 20 or 30 seconds
without headphones, and after that he has to place the headphones on and play the music
in his music player for the control volume adjustment.
• Remain to the subject to contact with the text experimenter after each control volume
adjustment by means of the inter-communicator.
• Explain to the subject how the inter-communicator system works.
• Remain to the subject that the first control volume adjustment does not count as data for
the listening test. It is only in order to make him used to the task.
• Tell the subject that there is no wrong answers since we are interested in his/her opinion.
• When the subject finishes each control volume adjustment, go back into the listening room
and conduct him to the waiting room.
• Then go into the listening room again and measure the SPL in the head and torso simulator
produced by the music player of the subject, using the music sample and the control
volume position that he selected. Do not forget to save the data in the measurement
system.
• When all the volume adjustments have finished tell the subject that the listening test is
over.
• Answer possible questions that the subject can have.
• Say “Thank you”, and show the subject to the entrance.
103
APPENDIX I. INSTRUCTIONS FOR THE TEST EXPERIMENTER
104
A PPENDIX J
L ISTENING T EST Q UESTIONNAIRE
J.0.5 Personal data:
• Name and Surname:...............................................................................................
• Music Player:........................................................................................................
• Headphones:.........................................................................................................
• Music Sample:.........................................................................................................
J.0.6 Prior experiene:
• Are you suffering or have you had recently a cold?
Yes
No
I dont’t Know
• Have you ever taken any medicine or another type of drugs that might have influence on
your hearing?
Yes
No
I dont’t Know
• Have you ever had?
105
APPENDIX J. LISTENING TEST QUESTIONNAIRE
buzzing noises in your ear
sudden lost of hearing
wax in your ears
pain or headache due to powerful sounds
hearing problems when you are listening TV
hearing problems when you talk by phone
nervous or anxious feeling after being in a noisy area
ringing in the ears after being in a noisy area
pain in your ears when you travel by plane
hearing problems after listening to music
using portable music players with headphones
Never
Sometimes
Often
Very often
J.0.7 Listening habits:
• Please, fill next table with the corresponding time that you use to listen to your music
player with headphones in quiet/moderate environments such as the street, the university
or at home for example:
How often?
weekly
monthly
Day
1
2
3
4
5
6
7
Time in hours per day
1 or less
1-2
2-4
4-6
6-8
8-10
more than 10
How long?
1 month
1-6 months
6-12 months
1-2 years
2-5 years
5-10 years
more than 10 years
• Please, fill next table with the corresponding time that you use to listen to your music
player with headphones in noisy environments such as the bus, the train or the bike for
example:
How often?
weekly
monthly
Day
1
2
3
4
5
6
7
Time in hours per day
1 or less
1-2
2-4
4-6
6-8
8-10
more than 10
Thank you for your participation.
Group 1066. Acoustic Department.
106
How long?
1 month
1-6 months
6-12 months
1-2 years
2-5 years
5-10 years
more than 10 years
A PPENDIX K
L ISTENING T EST S UBJECTS
This appendix lists the subjects who performed the listening test and their results.
According to the results of the pilot test, some subjects are classified as target and some as control population.
The subjects of the control and target population who performed the listening test are shown in
Table K.1 and K.2 respectively. The reference number and an assigned nickname is specified
in this table to differ among subjects. Moreover, this table illustrate the LEX,8h calculated from
the listening test data as well as the LFF,Aeq for the different environments tested (very quiet,
quiet/moderate, noisy).
Figures K.1 and K.2 show the LFF,Aeq and the exposure time for each subject in both populations
and for very quiet, quiet/moderate and noisy environment. The data from these two graph can be
combined to obtain the LEX,8h , which is show in Figure K.3, for each subject in both populations
too.
Figure K.4 shows the mean LF F,Aeq for both populations in the different environments tested
(very quiet, quiet/moderate, noisy). Moreover, Figure K.5 shows the mean LEX,8h for a quiet/moderate and noisy environment.
Figures K.6, K.7 and K.8 illustrate again the data of the LFF,Aeq selected for all the subject in
both populations, but as a function of the environments.
Figure K.9 shows the information obtained from the prior experience of the listening test questionnaire.
107
APPENDIX K. LISTENING TEST SUBJECTS
Reference Number
Table K.1:
Nickname
2
VLL
3
DVD
5
HDA
7
MCK
8
ANT
12
DNL
13
BTZ
15
LRM
17
IRS
18
HCR
LFF,Aeq
and
LEX,8h
Environment
Very Quiet
Quiet/Moderate
Noisy
Very Quiet
Quiet/Moderate
Noisy
Very Quiet
Quiet/Moderate
Noisy
Very Quiet
Quiet/Moderate
Noisy
Very Quiet
Quiet/Moderate
Noisy
Very Quiet
Quiet/Moderate
Noisy
Very Quiet
Quiet/Moderate
Noisy
Very Quiet
Quiet/Moderate
Noisy
Very Quiet
Quiet/Moderate
Noisy
Very Quiet
Quiet/Moderate
Noisy
LF F,Aeq
73,94
90,21
100,80
85,28
93,32
99,45
98,13
100,36
98,54
88,50
90,78
91,86
102,04
98,88
102,04
76,95
88,39
99,53
77,56
91,56
92,99
83,34
89,56
95,61
97,73
103,34
98,98
94,80
94,79
95,36
Hours per Day
LEX,8h
0,00
0,00
0,00
0,00
0,00
0,00
0,00
0,00
0,00
0,00
0,00
0,00
0,00
0,00
0,00
0,00
0,00
0,00
0,00
0,00
0,67
0,67
77,60
88,74
0,80
0,13
81,56
75,21
0,53
0,03
77,80
71,81
0,67
0,27
92,55
84,21
0,80
0,03
84,79
71,56
alulated for the subjets of the ontrol population aording to the data
obtained in the listening test for a very quiet, a quiet/moderate and a noisy environment.
108
Reference Number
Table K.2:
Nickname
31
LRI
33
CLS
34
ERC
39
SMN
41
YSN
42
NCL
45
SUS
46
OLV
50
JON
52
CLA
LFF,Aeq
and
LEX,8h
Environment
Very Quiet
Quiet/Moderate
Noisy
Very Quiet
Quiet/Moderate
Noisy
Very Quiet
Quiet/Moderate
Noisy
Very Quiet
Quiet/Moderate
Noisy
Very Quiet
Quiet/Moderate
Noisy
Very Quiet
Quiet/Moderate
Noisy
Very Quiet
Quiet/Moderate
Noisy
Very Quiet
Quiet/Moderate
Noisy
Very Quiet
Quiet/Moderate
Noisy
Very Quiet
Quiet/Moderate
Noisy
LF F,Aeq
90,74
96,04
103,26
102,56
104,82
107,59
94,80
97,30
103,32
99,10
91,56
97,57
88,53
97,49
93,56
87,86
92,43
97,27
100,16
103,24
105,11
103,02
98,83
103,91
102,07
103,08
106,30
104,12
103,55
96,07
Hours per Day
LEX,8h
0,03
0,47
72,24
90,92
1,07
0,07
96,07
86,79
0,93
0,40
87,97
90,31
0,67
0,93
80,77
88,24
0,67
2,67
86,70
88,79
0,03
1,07
68,63
88,52
0,03
0,27
79,44
90,34
1,87
0,13
92,51
86,13
0,17
1,33
86,27
98,52
1,33
1,33
95,76
88,28
measured in dB SPL alulated for the subjets of the target population
aording to the data obtained in the listening test for a very quiet, a quiet/moderate and a noisy
environment.
109
APPENDIX K. LISTENING TEST SUBJECTS
Control Population
LFF,Aeq [dB SPL]
110
100
90
80
70
VLL
DVD
HDA
MCK
ANT
DNL
BTZ
LRM
IRS
HCR
Very quiet Environment
Quiet/moderate Environment
Noisy Environment
Target Population
LFF,Aeq [dB SPL]
110
100
90
80
70
Figure K.1:
LFF,Aeq
LRI
CLS
ERC
SMN
YSN
NCL
Subjects
SUS
OLV
JON
CLA
alulated for the subjets of the target and ontrol population aording to the data
obtained in the listening test for a very quiet, a quiet/moderate and a noisy environment.
Control Population
3
Time [hours/day]
2.5
2
1.5
1
0.5
0
VLL
DVD
HDA
MCK
ANT
DNL
BTZ
LRM
IRS
HCR
Very quiet Environment
Quiet/moderate Environment
Noisy Environment
Target Population
3
Time [hours/day]
2.5
2
1.5
1
0.5
0
LRI
CLS
ERC
SMN
YSN
NCL
Subjects
SUS
OLV
JON
CLA
Figure K.2: Exposure time alulated for the subjets of the target and ontrol population aording to the
data obtained in the listening test for a quiet/moderate and a noisy environment.
110
Control Population
Lex,8h [dB SPL]
100
90
80
70
60
VLL
DVD
HDA
MCK
ANT
DNL
BTZ
LRM
IRS
HCR
Quiet/moderate Environment
Noisy Environment
Target Population
Lex,8h [dB SPL]
100
90
80
70
60
Figure K.3:
LRI
LEX,8h
CLS
ERC
SMN
YSN
NCL
Subjects
SUS
OLV
JON
CLA
alulated for the subjets of the target and ontrol population aording to the data
obtained in the listening test for a quiet/moderate and a noisy environment.
111
APPENDIX K. LISTENING TEST SUBJECTS
110
Very quiet Environment
Quiet/moderate Environment
Noisy Environment
Mean LFF,Aeq [dB SPL]
105
100
95
90
85
80
Control
Target
Populations
Figure K.4: Mean
LFF,Aeq
alulated aross all the subjets of the target and ontrol population aording to
the data obtained in the listening test for a very quiet, a quiet/moderate and a noisy environment.
110
Quiet/moderate Environment
Noisy Environment
100
Mean Lex,8h [dB SPL]
90
80
70
60
50
Control
Target
Populations
Figure K.5: Mean
LEX,8h
alulated aross all the subjets of the target and ontrol population aording to
the data obtained in the listening test for a quiet/moderate and a noisy environment.
112
105
VLL
DVD
HDA
MCK
DNL
LRM
IRS
BTZ
ANT
HCR
100
L FF,Aeq [dB SPL]
95
90
85
80
75
70
very quiet
quiet/moderate
noisy
Environments
Figure K.6:
LFF,Aeq
alulated for the subjets of the ontrol population aording to the data obtained in
the listening test for a very quiet, a quiet/moderate and a noisy environment.
110
LRI
CLS
ERC
SMN
YSN
NCL
SUS
OLV
JON
CLA
L FF,Aeq [dB SPL]
105
100
95
90
85
very quiet
quiet/moderate
noisy
Environments
Figure K.7:
LFF,Aeq
alulated for the subjets of the target population aording to the data obtained in the
listening test for a very quiet, a quiet/moderate and a noisy environment.
113
APPENDIX K. LISTENING TEST SUBJECTS
105
LFF,Aeq Mean Control Population
LFF,Aeq Mean Target Population
LFF.Aeq [dB SPL]
100
95
90
85
very quiet
quiet/moderate
noisy
Environments
Figure K.8: Mean
LFF,Aeq
alulated aross all the subjets of the target and ontrol population aording to
the data obtained in the listening test for a very quiet, a quiet/moderate and a noisy environment.
Control Population
Number of subjects
10
8
6
4
2
0
1
2
3
4
5
6
7
8
9
10
Never
Sometimes
Often
Very Often
Target Population
Number of subjects
10
8
6
4
2
0
1
2
3
4
5
6
7
Prior experience items
8
9
10
Figure K.9: Number of people from the ontrol and target population who suer: 1-buzzing noises in your
ear, 2-sudden lost of hearing, 3-wax in your ears, 4-pain due to powerful sounds, 5-hearing
problems when listening TV, 6-hearing problems when talking by phone, 7-nervous after being
in a noisy area, 8-ringing in the ears after being in a noisy area, 9-pain the ears when traveling
by plane,10-hearing problems after listening a MP with headphones.
114
A PPENDIX L
H EARING T HRESHOLD AND DPOAE S
This appendix describes the pure tone audiometry and the DPOAEs measurements conducted
in the listening test in order to assess the hearing of the subjects. The appendix finishes with the
data obtained from these measurements.
L.1 Pure Tone Audiometry
The audiometry is carried out in the audiometric cabin B4 from the Acoustics Department at
Aalborg University, which is a room with special treatment to the walls, ceiling, and floor in
order to comply with the specifications about the noise floor stated in the standard ISO 8253-1
[26].
The audiometer Madser Orbiter 922 is chosen to perform the audiometry test. The audiometry
method used is the pure tone audiometry, which is selected in Section 2.7 in Chapter 2. The
subject is equipped with a set of headphones and two roller pens. One of these roller pens is red
and the other one is blue corresponding to right and left respectively. The device tests the ears
one at a time. The subject must press the corresponding roller pens whenever a sound is heard.
In the end of this Appendix the results obtained in the audiometry tests are shown.
L.2 DPOAEs Measurements
The measurements are performed in the audiometer cabin B4 in the Acoustic Department at
Aalborg University where there was a noise controlled environment. The equipment used to
carry out the DPOAEs measurements is the ILO96 from Otodynamic. It is selected since it is
the equipment available in the Acoustic Department at Aalborg University. This equipment can
measure either TEOAEs or DPOAEs.
The equipment is based on a system amplification unit connected to a computer which has installed the software to measure and record the OAEs. The amplification unit has, apart from the
connection to the computer, an ear probe which contains two small loudspeakers and a microphone. This ear probe is placed on the ear of the subject using a small foam which makes the
placement on the ear easier. Every loudspeaker of the probe is used to play each pure tone and
115
APPENDIX L. HEARING THRESHOLD AND DPOAES
the microphone is used to record the DPOAEs produced by the ear of the subject.
DPOAEs were measured for both ears in each subject. The data from the measurement is collected in .SPR format and it is later processed in order to plot the results. The results obtained
from these measurements are presented in the last section of this appendix.
After each measurement, the level of the DPOAEs is compared to the noise level to verify it
is not masking the cochlear emissions. This can be seen in the equipment immediately after
the measurement is carried out. If noise level masks the cochlear emissions, the measurements
are repeated. Moreover in the ear calibration is done for each subject and for each ear. This is
automatically performed by the DPOAEs measurement system.
Not only external noise can alters the OAEs response, but also the internal noises from the subject (coughing for example) can create difficulties when recording the DPOAEs. Therefore it is
important to ask for the subject to remain quiet and try to not swallow during the measurements.
L.3 Results
The data obtained in the audiometry test was collected and is shown in Table L.2 and L.1.
The data obtained from the DPOAEs measurements is presented in one plot for every subject.
The DPOEAs of the right and left ears are presented in the same graph, but in different colors.
The thick red line in each graph correspond to DPOAEs or the right ear, whereas the thick blue
line represent the results of the DPOAEs measurements of the left ear. The noise measured,
which is plotted as blue and red fine lines in the graphs, corresponds to the noise measured
during the DPOAEs measurements for the right and left ears respectively. The DPOAEs data is
√
plotted as a function of the geometric mean of the pure tones of the DPOAEs ( f1 . f 2 ).
116
L.3. RESULTS
Subject
VLL
DVD
HDA
MCK
ANT
DNL
BTZ
LRM
IRS
HCR
Ear
L
R
L
R
L
R
L
R
L
R
L
R
L
R
L
R
L
R
L
R
250
-5
-5
0
10
5
5
5
0
0
0
5
0
5
5
5
-5
0
-5
-5
0
500
5
5
5
5
5
10
15
10
5
5
5
5
5
0
5
0
0
0
5
5
750
5
10
0
5
5
0
20
10
0
0
5
5
10
5
0
-5
-5
5
10
5
1000
5
10
5
5
5
0
15
5
5
0
0
0
5
0
5
0
-10
0
10
5
Frequencies [Hz]
1500 2000 3000
5
5
0
5
10
0
5
10
5
5
5
0
5
5
0
0
5
0
15
15
10
5
-5
5
-5
0
0
0
0
0
5
0
0
5
0
-5
5
0
-5
0
0
0
5
5
0
5
10
0
0
5
0
15
5
5
5
5
10
10
10
10
4000
0
0
0
5
5
15
25
5
10
10
5
10
0
5
5
0
15
0
15
15
6000
15
0
10
10
15
25
20
5
5
-5
20
25
15
15
15
10
5
20
10
5
8000
0
-10
-10
5
-5
5
10
0
-10
-5
0
10
5
5
5
0
-5
-5
5
-10
Table L.1: Hearing thresholds of the test subjets of the ontrol population measured in the listening test.
−15
Mean Right Ear Target Population
Mean Left Ear Target Population
−10
Threshold [dB HL]
−5
0
5
10
15
20
25
250
500
750
1k
1.5 k
2k
Frequency [Hz]
3k
4k
6k
8k
Figure L.1: Mean and standard deviation of the hearing thresholds of the target population for left and right
ear.
117
APPENDIX L. HEARING THRESHOLD AND DPOAES
Subject
EDR
LRI
CLS
ERC
JLS
SMN
YSN
NCL
SUS
OLV
JON
CLA
Ear
L
R
L
R
L
R
L
R
L
R
L
R
L
R
L
R
L
R
L
R
L
R
L
R
250
10
0
-10
-10
-5
5
5
5
10
5
5
10
5
5
-5
5
5
0
5
0
5
-10
10
20
500
10
10
-10
-10
-5
5
5
5
5
10
0
5
0
0
-5
0
0
0
5
0
5
0
10
15
750
5
5
5
-5
5
-5
15
5
5
15
5
0
5
-10
-5
5
0
-5
0
5
5
0
5
5
1000
5
-5
5
0
-5
-5
10
5
5
20
0
0
5
0
0
0
0
0
0
5
0
5
0
5
Frequencies [Hz]
1500 2000 3000
10
5
15
-10
0
0
-10
5
0
-10
-10
0
-5
5
5
10
5
5
5
5
0
0
0
5
15
10
15
25
25
20
5
5
0
5
5
15
-10
-10
-10
5
-10
-10
5
0
5
5
0
0
5
5
-5
0
5
-5
-10
-5
-10
0
-10
-10
0
0
-5
5
5
-5
5
0
0
10
5
5
4000
35
5
-10
-10
-5
5
5
10
15
20
5
5
5
-5
15
-5
5
0
0
-10
5
-10
5
-5
6000
50
-10
10
10
10
-5
0
0
15
20
10
15
15
20
10
5
5
5
10
-5
20
15
10
5
8000
45
5
5
5
-5
0
-5
0
10
35
5
10
-10
15
0
-10
-5
-5
-10
-10
0
-5
-10
-10
Table L.2: Hearing thresholds of the test subjets of the target population measured in the listening test.
−15
Mean Right Ear Control Population
Mean Left Ear Control Population
−10
Threshold [dB HL]
−5
0
5
10
15
20
25
250
500
750
1k
1.5 k
2k
Frequency [Hz]
3k
4k
6k
8k
Figure L.2: Mean and standard deviation of the hearing thresholds of the ontrol population for left and right
ear.
118
L.3. RESULTS
DPOAE: Subject VLL, Right and Left ears
30
25
DPOAE Right Ear
DPOAE Left Ear
Noise measured Right Ear
Noise measured Left Ear
20
DPOAE [dB SPL]
15
10
5
0
−5
−10
−15
−20
1000
1500
2000
3000
4000
6000
Frequency [Hz]: geometric mean
Figure L.3: DPOAEs of right and left ears in subjet VLL. This subjet belongs to the ontrol population.
DPOAE: Subject DVD, Right and Left ears
30
25
DPOAE Right Ear
DPOAE Left Ear
Noise measured Right Ear
Noise measured Left Ear
20
DPOAE [dB SPL]
15
10
5
0
−5
−10
−15
−20
1000
1500
2000
3000
4000
6000
Frequency [Hz]: geometric mean
Figure L.4: DPOAEs of right and left ears in subjet DVD. This subjet belongs to the ontrol population.
119
APPENDIX L. HEARING THRESHOLD AND DPOAES
DPOAE: Subject HDA, Right and Left ears
30
25
DPOAE Right Ear
DPOAE Left Ear
Noise measured Right Ear
Noise measured Left Ear
20
DPOAE [dB SPL]
15
10
5
0
−5
−10
−15
−20
1000
1500
2000
3000
4000
6000
Frequency [Hz]: geometric mean
Figure L.5: DPOAEs of right and left ears in subjet HDA. This subjet belongs to the ontrol population.
DPOAE: Subject MCK, Right and Left ears
30
25
DPOAE Right Ear
DPOAE Left Ear
Noise measured Right Ear
Noise measured Left Ear
20
DPOAE [dB SPL]
15
10
5
0
−5
−10
−15
−20
1000
1500
2000
3000
4000
6000
Frequency [Hz]: geometric mean
Figure L.6: DPOAEs of right and left ears in subjet MCK. This subjet belongs to the ontrol population.
120
L.3. RESULTS
DPOAE: Subject ANT, Right and Left ears
30
25
DPOAE Right Ear
DPOAE Left Ear
Noise measured Right Ear
Noise measured Left Ear
20
DPOAE [dB SPL]
15
10
5
0
−5
−10
−15
−20
1000
1500
2000
3000
4000
6000
Frequency [Hz]: geometric mean
Figure L.7: DPOAEs of right and left ears in subjet ANT. This subjet belongs to the ontrol population.
DPOAE: Subject DNL, Right and Left ears
30
25
DPOAE Right Ear
DPOAE Left Ear
Noise measured Right Ear
Noise measured Left Ear
20
DPOAE [dB SPL]
15
10
5
0
−5
−10
−15
−20
1000
1500
2000
3000
4000
6000
Frequency [Hz]: geometric mean
Figure L.8: DPOAEs of right and left ears in subjet DNL. This subjet belongs to the ontrol population.
121
APPENDIX L. HEARING THRESHOLD AND DPOAES
DPOAE: Subject BTZ, Right and Left ears
30
25
DPOAE Right Ear
DPOAE Left Ear
Noise measured Right Ear
Noise measured Left Ear
20
DPOAE [dB SPL]
15
10
5
0
−5
−10
−15
−20
1000
1500
2000
3000
4000
6000
Frequency [Hz]: geometric mean
Figure L.9: DPOAEs of right and left ears in subjet BTZ. This subjet belongs to the ontrol population.
DPOAE: Subject LRM, Right and Left ears
30
25
DPOAE Right Ear
DPOAE Left Ear
Noise measured Right Ear
Noise measured Left Ear
20
DPOAE [dB SPL]
15
10
5
0
−5
−10
−15
−20
1000
1500
2000
3000
Frequency [Hz]: geometric mean
4000
6000
Figure L.10: DPOAEs of right and left ears in subjet LRM. This subjet belongs to the ontrol population.
122
L.3. RESULTS
DPOAE: Subject IRS, Right and Left ears
30
25
DPOAE Right Ear
DPOAE Left Ear
Noise measured Right Ear
Noise measured Left Ear
20
DPOAE [dB SPL]
15
10
5
0
−5
−10
−15
−20
1000
1500
2000
3000
4000
6000
Frequency [Hz]: geometric mean
Figure L.11: DPOAEs of right and left ears in subjet IRS. This subjet belongs to the ontrol population.
DPOAE: Subject HCR, Right and Left ears
30
25
DPOAE Right Ear
DPOAE Left Ear
Noise measured Right Ear
Noise measured Left Ear
20
DPOAE [dB SPL]
15
10
5
0
−5
−10
−15
−20
1000
1500
2000
3000
4000
6000
Frequency [Hz]: geometric mean
Figure L.12: DPOAEs of right and left ears in subjet HCR. This subjet belongs to the ontrol population.
123
APPENDIX L. HEARING THRESHOLD AND DPOAES
DPOAE: Subject EDR, Right and Left ears
30
25
DPOAE Right Ear
DPOAE Left Ear
Noise measured Right Ear
Noise measured Left Ear
20
DPOAE [dB SPL]
15
10
5
0
−5
−10
−15
−20
1000
1500
2000
3000
4000
6000
Frequency [Hz]: geometric mean
Figure L.13: DPOAEs of right and left ears in subjet EDR. This subjet belongs to the target population.
DPOAE: Subject LRI, Right and Left ears
30
25
DPOAE Right Ear
DPOAE Left Ear
Noise measured Right Ear
Noise measured Left Ear
20
DPOAE [dB SPL]
15
10
5
0
−5
−10
−15
−20
1000
1500
2000
3000
4000
6000
Frequency [Hz]: geometric mean
Figure L.14: DPOAEs of right and left ears in subjet LRI. This subjet belongs to the target population.
124
L.3. RESULTS
DPOAE: Subject CLS, Right and Left ears
30
25
DPOAE Right Ear
DPOAE Left Ear
Noise measured Right Ear
Noise measured Left Ear
20
DPOAE [dB SPL]
15
10
5
0
−5
−10
−15
−20
1000
1500
2000
3000
4000
6000
Frequency [Hz]: geometric mean
Figure L.15: DPOAEs of right and left ears in subjet CLS. This subjet belongs to the target population.
DPOAE: Subject ERC, Right and Left ears
30
25
DPOAE Right Ear
DPOAE Left Ear
Noise measured Right Ear
Noise measured Left Ear
20
DPOAE [dB SPL]
15
10
5
0
−5
−10
−15
−20
1000
1500
2000
3000
Frequency [Hz]: geometric mean
4000
6000
Figure L.16: DPOAEs of right and left ears in subjet ERC. This subjet belongs to the target population.
125
APPENDIX L. HEARING THRESHOLD AND DPOAES
DPOAE: Subject LRM, Right and Left ears
30
25
DPOAE Right Ear
DPOAE Left Ear
Noise measured Right Ear
Noise measured Left Ear
20
DPOAE [dB SPL]
15
10
5
0
−5
−10
−15
−20
1000
1500
2000
3000
4000
6000
Frequency [Hz]: geometric mean
Figure L.17: DPOAEs of right and left ears in subjet JLS. This subjet belongs to the target population.
DPOAE: Subject SMN, Right and Left ears
30
25
DPOAE Right Ear
DPOAE Left Ear
Noise measured Right Ear
Noise measured Left Ear
20
DPOAE [dB SPL]
15
10
5
0
−5
−10
−15
−20
1000
1500
2000
3000
4000
6000
Frequency [Hz]: geometric mean
Figure L.18: DPOAEs of right and left ears in subjet SMN. This subjet belongs to the target population.
126
L.3. RESULTS
DPOAE: Subject YSN, Right and Left ears
30
25
DPOAE Right Ear
DPOAE Left Ear
Noise measured Right Ear
Noise measured Left Ear
20
DPOAE [dB SPL]
15
10
5
0
−5
−10
−15
−20
1000
1500
2000
3000
4000
6000
Frequency [Hz]: geometric mean
Figure L.19: DPOAEs of right and left ears in subjet YSN. This subjet belongs to the target population.
DPOAE: Subject NCL, Right and Left ears
30
25
DPOAE Right Ear
DPOAE Left Ear
Noise measured Right Ear
Noise measured Left Ear
20
DPOAE [dB SPL]
15
10
5
0
−5
−10
−15
−20
1000
1500
2000
3000
4000
6000
Frequency [Hz]: geometric mean
Figure L.20: DPOAEs of right and left ears in subjet NCL. This subjet belongs to the target population.
127
APPENDIX L. HEARING THRESHOLD AND DPOAES
DPOAE: Subject SUS, Right and Left ears
30
25
DPOAE Right Ear
DPOAE Left Ear
Noise measured Right Ear
Noise measured Left Ear
20
DPOAE [dB SPL]
15
10
5
0
−5
−10
−15
−20
1000
1500
2000
3000
4000
6000
Frequency [Hz]: geometric mean
Figure L.21: DPOAEs of right and left ears in subjet SUS. This subjet belongs to the target population.
DPOAE: Subject OLV, Right and Left ears
30
25
DPOAE Right Ear
DPOAE Left Ear
Noise measured Right Ear
Noise measured Left Ear
20
DPOAE [dB SPL]
15
10
5
0
−5
−10
−15
−20
1000
1500
2000
3000
4000
6000
Frequency [Hz]: geometric mean
Figure L.22: DPOAEs of right and left ears in subjet OLV. This subjet belongs to the target population.
128
L.3. RESULTS
DPOAE: Subject JON, Right and Left ears
30
25
DPOAE Right Ear
DPOAE Left Ear
Noise measured Right Ear
Noise measured Left Ear
20
DPOAE [dB SPL]
15
10
5
0
−5
−10
−15
−20
1000
1500
2000
3000
4000
6000
Frequency [Hz]: geometric mean
Figure L.23: DPOAEs of right and left ears in subjet JON. This subjet belongs to the target population.
DPOAE: Subject CLA, Right and Left ears
30
25
DPOAE Right Ear
DPOAE Left Ear
Noise measured Right Ear
Noise measured Left Ear
20
DPOAE [dB SPL]
15
10
5
0
−5
−10
−15
−20
1000
1500
2000
3000
4000
6000
Frequency [Hz]: geometric mean
Figure L.24: DPOAEs of right and left ears in subjet CLA. This subjet belongs to the target population.
129
APPENDIX L. HEARING THRESHOLD AND DPOAES
30
Mean DPOAE Left Ear Control Population
Mean DPOAE Right Ear Control Population
Mean Noise DPOAE Left Ear Control Population
Mean Noise DPOAE Right Ear Control Population
25
20
DPOAE [dB SPL]
15
10
5
0
−5
−10
−15
−20
1000
1500
2000
3000
Frequency [Hz]: geometric mean
4000
6000
Figure L.25: Mean and standard deviation DPOAE of the test subjets of the ontrol population for left and
right ear.
30
Mean DPOAE Left Ear Target Population
Mean DPOAE Right Ear Target Population
Mean Noise DPOAE Left Ear Target Population
Mean Noise DPOAE Right Ear Target Population
25
20
DPOAE [dB SPL]
15
10
5
0
−5
−10
−15
−20
1000
1500
2000
3000
Frequency [Hz]: geometric mean
4000
6000
Figure L.26: Mean and standard deviation DPOAE of the test subjets of the target population for left and
right ear.
130
A PPENDIX M
L ISTENING E NVIRONMENTS
The following appendix describes the simulation of the quiet/moderate and noisy environments
of the listening test. The background noise in a bus and in a street are used as references
environments to simulate these situations respectively.
M.1 Method
The simulation is done in a listening room by playing in an omnidirectional source two different
signals corresponding to each environment. The omnidirectional sound source is used because
it allows to create a semi-diffuse sound field where the sound pressure level is approximately the
same in all the room. This is desired because in a real situation where a MP with headphones is
used, the background noise generally corresponds to a semi-diffuse sound field.
M.2 Equipment and Setup
The complete listening test setup is shown in Figure M.1. In this figure it can be seen the sound
source which produces the environments as well as the computer and the amplifier that drive the
signals. Table M.1 lists the equipment utilized.
Item
Omnidirectional source
Power Amplifier
Computer
Model
LAB-Nr:
00000000
00000000
00000000
00000000
Table M.1: Equipment used for the listening test.
M.2.1 Sound Stimuli for the Environments
Most noise in the environments contains energy at many different frequencies combining together to give it its overall character. Therefore is complicated to analyze these environments.
The noise from a bus is a continuous non stationary noise, which has its energy content at low
frequency. Moreover the amplitude of the signal varies over the time because the noise emitted
131
APPENDIX M. LISTENING ENVIRONMENTS
in
Poweramp
in
out
Omnidirectional
source
test subject
PC
out
Listening room
Figure M.1: Setup used for the listening test.
The subjet is equipped with his own portable MP or with
the referene portable MP used in the pilot test. The task of the subjet in this part of the
listening test is to adjust the ontrol volume position when a noisy (bus), a moderate (street),
and a quiet (silene) sound environments are being simulated.
by a bus depends on its speed, its acceleration and many other parameters. Therefore it is a
complex sound which is difficult to simulate.
The noise from a street can be related to many different noises. In this case it is defined as
environmental noise, for example the song of a bird, some cars passing by or the sound of the
wind, corresponding to a quiet or moderate environment.
These environments are simulated with two WAV files: one recorded inside of a bus and one
recorded in a street. These signals are obtained from a sound database in Internet [2]. The signal
for the bus environment is a sample of 46 seconds which contains some of the different noises
that can be produced in a bus, for example people talking, the sound produced when opening
and closing the doors of the bus, or the sound produced by the bus when braking or starting.
The signal for the street environment is a sample of 90 seconds which contains different noises
that can be recorded in a quiet street.
Figures M.3 and M.2 show the time signals that are used.
A possible problem with these signals is that they are not very long, so the subject may need
more time than the duration of these files to adjust the volume. For this reason each of these
signals is concatenated until getting a duration of 5 minutes approximately each one. The concatenation is done as a linear cross-fade of 5 seconds between the signals to concatenate. The
signals are amplified such that the energy of the end of the first file is equal to the energy of the
next file. Figure M.4 shows an example of a pair of files of the street environment concatenated.
132
M.2. EQUIPMENT AND SETUP
1
0.8
0.6
0.4
Amplitude
0.2
0
−0.2
−0.4
−0.6
−0.8
−1
0
5
10
15
20
25
30
35
40
45
Time(s)
Figure M.2: Noise signal used for the simulation of the
bus listening
environment
In order to decide the sound pressure level at which these files have to be played, some field
measurements are performed inside of two buses and in a street..
M.2.2 Results
Noise levels were measured using a Monacor SM-4 sound level meter set to a A-weighting slow
meter response. The slow meter response is selected because the reading of the values in the
sound level meter is easier. This is because the integration time in order to calculate the sound
pressure level is longer when using the slow meter response in the sound level meter.
The objective of these measurements is to obtain a maximum and a minimum sound pressure
level for each of the environments mentioned before. In order to do that several measurements
are done in each environment. The time of the performance of the measurement is always
between 10 A.M. and 4 P.M.
M.2.3 Bus Measurements
These measurements are performed inside of the buses number 2 and number 12 at Aalborg
(Denmark). Measurements in bus stops are not carried out.
Since the noise in a bus may vary depending on many variables such as the recording time,
the type of bus, the traffic noise, the number of passengers traveling in the bus, etc. a total of
18 measurements are performed in different moments. These measurements are carried out in
three different positions in the bus: In the end, in the middle and in the beginning of the bus.
Moreover two repetitions are performed in each position in different moments: When the bus
stops, when the bus moves and when the bus starts or brakes.
133
APPENDIX M. LISTENING ENVIRONMENTS
1
0.8
0.6
0.4
Amplitude
0.2
0
−0.2
−0.4
−0.6
−0.8
−1
0
10
20
30
40
50
60
70
80
90
Time(s)
Figure M.3: Noise signal used for the simulation of the
street
listening environment
M.2.4 Street Measurements
To estimate the sound pressure level in a street is not a easy task, as well as in the bus, because
many parameters have an important influence in the sound pressure level measured. Since the
idea is to simulate an environment with a moderate or low noise level, a not very noisy street
close to Aalborg University is used. The street selected is Fredrik Bajers Vej. The measurements
are performed close to the entrance of the building seven of Aalborg University, which is in this
street. There are different noises that can be recorded in this street depending on the situation.
In this case the measurement is focus on recording the level of the sound produced by people
talking in the street with a moderate or low environmental noise. Moreover the position in the
street is taken into account for the recordings. Therefore two positions are defined: close to the
road and far away from the road. A total of 9 measurements are performed in different moments
depending on the amount of people present.
M.2.5 Field Measurements Results
The data were collected by a reading of the screen in the sound level meter. Table M.2 and Table
M.3 shows the data collected for the street environment and the bus environment respectively.
These tables show A-weighted sound pressure levels.
From the results of this appendix, the maximum and minimum sound pressure level measured
for the listening environments are obtained. These are 90 and 106.2 dB for the bus listening
environment and 64.5 and 77, 3 dB SPL for the street listening environment.
After testing these levels in the listening room used for the simulation it was observed that the
noise levels in the bus situation where to high compare to a read situation. For these reason it
was decided to reduce this levels until a realistic situation was achieved.
134
M.2. EQUIPMENT AND SETUP
1
0.8
0.6
0.4
Amplitude
0.2
0
−0.2
−0.4
−0.6
−0.8
−1
Figure M.4: Signal of the
street
20
40
60
80
Time(s)
100
120
140
160
listening environment reated by onatenating the seleted le twie in a
row.
State of the bus
stopped
starting
moving
stopped
starting
moving
Positions in the bus
Beginning Middle
End
90
88,8
95
93,7
96,0
98
96
95,7
106,2
89,0
90
97
94,4
94
98,8
99,7
98
104,7
Table M.2: A-weighted sound pressure level measured for the
bus listening environment.
Three dierent posi-
tions in the bus and three dierent states of the bus are taken into aount. These measurements
are performed twie.
Position in the street
close to the road
far from the road
close to the road
far from the road
Number of people present
0-2 persons More than 2 persons
65,4
70,2
50,9
63,9
77,3
70,9
66,0
65,4
Table M.3: A-weighted sound pressure level measured for the
street
listening environment. Dierent situa-
tions are taken into aount depending on the amount of people present in the reording and
depending on the position in the street. These measurements are performed twie.
135
A PPENDIX N
AN ALISY O F VA RIANCE (ANOVA)
In this appendix it can be found a brief explanation of the ANOVA test method applied.
ANOVA is used to test for the equality of several means. The test designed with two factor
requires a two-way ANOVA. By which it is possible to test the equality among the treatment
means of the two factors as well as the interactions between them. The treatments means of each
factor is tested for equality and it can be stated if, they are different within a level of significance.
Furthermore, the two-way ANOVA provides a knowledge about the interaction effect of the factors. This effect is best explained by a small example: A person might like both fish and meat
separately. But when mixing them (interaction), the taste might not please the person.
Statistical hypotheses are a statement about the problem situation. The population mean for the
i’th treatment is denoted µi , the tested null-hyphothesis H0 is denoted as:
H0 : µ1 = µ2 = ... = µi .
In order to test the hyphoteses the standard level of significance among statisticians is used. This
value is noted as α, moreover it is equal to 0.05. This implies a maximun of 5% of probability
of making a Type I Error. This error occurrs if the null hypothesis is rejected when it is true. It
can be noted as:
α = P(type I error) = P(re ject H0 |H0
is true)
Three assumptions are required for the ANOVA method to be exact, however a violations of
these assuptions does not necessarily to lead false conclusions:
1. The treatment data must be normally distributed
2. The variance must be the same for all treatments
3. The observations must be independient. Being an observation one of the possible convinations between the treatments of the two factors.
Figure N.1shows the normalization plot. The observations marked with circles are normally
distributed if the residual from the line is small.
137
APPENDIX N. ANALISY OF VARIANCE (ANOVA)
Figure N.1: Representation of the normalization plot of the dependent variables:
LFF,Aeq (left-up), LEX,8h
(right up), HL(left-down), DPOAEL(right-down)
The compliance with the second assumption can be appreciated from the standard deviation
values shown in the listening level, exposure level, hearing thresholds and DPOAEs analysis
performed for this project.
The third assumption can be assumed since there are two different groups of test subjects.
138
A PPENDIX O
I NDEPENDENT S AMPLES T-T EST
A NALYSIS
In this appendix it can be found a brief explanation of the T-tests test method applied.
An Independent Samples T-test compares the mean scores of two groups on a given variable. In
this test the dependent variables are: LFF,Aeq , HL and DPOAEL.
The next assumptions are made:
• The dependent variable is normally distributed. It can be appreciated from Figure N.1.
• The two groups have approximately equal variance on the dependent variable. The equality of variances is tested using Levenet’s Test. It is explained below.
• The two groups are independent of one another
The hypotheses formulated are:
• Null: The means of the two treatments tested are not significantly different.
• Alternate: The means of the two treatments tested are significantly different
The null hypotheses can be rejected if the p-value obtained from the t-test is less than 0.05.
O.0.6 Listening Level T-test
This section describes the T-Tests performed by the treatments defined for factor environment.
These treatments are: very quiet, quiet/moderate and noisy environment. The LF F,Aeq means are
compared in groups of two by two. Then, the three T-test performed are:
1. T-test beetween: Very quiet and quite/moderate environment
2. T-test beetween: Very quiet and noisy environment
3. T-test beetween: Quite moderate and noisy environment
139
APPENDIX O. INDEPENDENT SAMPLES T-TEST ANALYSIS
T-test between: very quiet and quite/moderate environment
In Table O.1 the results of the Levenet’s test are shown. The significance value (Sig) is 0.012
which is smaller than 0.05. Then, it can not be assumed that the variances of both groups are
approximately equal. Therefore a p-value equal to 0,084 can be read from the second line of
Table O.1.
HL
Assumption
Equal variances
Not Equal variances
Levene’s Test
F
Sig
6.933
0.012
t
df
-1.780
-1.780
42
33.980
t-test for Equality of Means
p-value
Mean
Std. Error
Difference Difference
0.082
-4.015
2.255
0.084
-4.015
2.255
Confidence Interval
Lower
Upper
-8.568
0.536
-8.600
0.568
Table O.1: Results from Independent Samples T-test between: very quiet and quite/moderate environment.
T-test between: quiet/moderate and noisy environment
Table O.2 shows the results of the Levenet’s test and the Independent Samples T-test. Since Sig
is equal to 0.186, it can assume that the variances are approximately equal.
The p-value obtained from the T-test is 0.067, which is greater than 0.05.
HL
Assumption
Equal variances
Not Equal variances
Levene’s Test
F
Sig
1,812
0.186
t
df
-1.880
-1.880
42
40.534
t-test for Equality of Means
p-value
Mean
Std. Error
Difference Difference
0.067
-2.788
1.482
0.067
-2.788
1.482
Confidence Interval
Lower
Upper
-5.780
0.204
-5.783
0.207
Table O.2: Results from Independent Samples T-test between: quiet/moderate and noisy environment.
T-test between: very quiet and noisy environment
Table O.3 shows that it can not be assumed equality of variances. However, the null hypothesis
can be rejected as the p-value is 0.004.
HL
Assumption
Equal variances
Not Equal variances
Levene’s Test
F
Sig
17.514
0.001
t
df
-3.148
-3.148
42
30.370
t-test for Equality of Means
p-value
Mean
Std. Error
Difference Difference
0.003
-6.804
2.162
0.004
-6.804
2.162
Confidence Interval
Lower
Upper
-11.165
-2.442
-11.215
-2.392
Table O.3: Results from Independent Samples T-test between: very quiet and noisy environment.
140
O.0.7 Hearing Threshold and DPOAE T-test
In this sections there is not described in detail the procedure followed to obtained the results
from each T-Tests performed. A single matrix which contains the p-value results from each test
is shown.
HL T-test among the ears of both populations
Table O.4 shows the p-values obtained from the HL T-Tests performed among the two ears of
both populations.
LC
LC
RC
LT
RT
0.378
0.000
0.000
RC
0.378
0.006
0.005
LT
0.000
0.006
RT
0.000
0.005
0.837
0.837
Table O.4: Results from Independent Samples T-test: among the ears of both populations
HL T-test among the frequenies
Table O.5 shows the p-values obtained from the HL T-Tests performed among the 10 frequencies
tested.
250
250
500
750
1000
1500
2000
3000
4000
6000
8000
0.287
0.309
0.537
0.410
0.641
0.783
0.211
0.000
0.037
500
0.287
1.000
0.571
0.843
0.546
0.169
0.679
0.000
0.002
750
0.309
1.000
0.534
0.851
0.565
0.190
0.689
0.000
0.003
1000
0.537
0.571
0.534
0.753
0.914
0.346
0.389
0.000
0.004
1500
0.410
0.843
0.851
0.753
0.705
0.266
0.578
0.000
0.005
2000
0.641
0.546
0.565
0.914
0.705
0.449
0.377
0.000
0.110
3000
0.783
0.169
0.190
0.346
0.266
0.449
0.134
0.000
0.620
4000
0.211
0.679
0.689
0.389
0.578
0.377
0.134
0.001
0.003
6000
0.000
0.000
0.000
0.000
0.000
0.000
0.000
0.001
8000
0.037
0.002
0.003
0.004
0.005
0.11
0.62
0.003
0.006
0.006
Table O.5: Results from Independent Samples T-test among the ears of both populations
DPOAE T-test among the frequenies
Table O.6 presents the p-values obtained from the DPOAE T-Tests performed among the two
ears of both populations. A dependence between right and left ears of control and target populations is found.
141
APPENDIX O. INDEPENDENT SAMPLES T-TEST ANALYSIS
LC
LC
RC
LT
RT
0.416
0.000
0.000
RC
0.416
0.000
0.001
LT
0.000
0.000
RT
0.000
0.001
0.568
0.568
Table O.6: Results from Independent Samples T-test among the ears of both populations
DPOAE T-test among the frequenies
The p-values obtained from the the DPOAE T-tests carried out among the 22 frequencies tested
in DPOAE are presented in table O.7.
142
905
905
984
1076
1172
1282
1398
1524
1658
1810
1980
2156
2350
2563
2793
3049
3316
3626
3948
4306
4695
5121
5582
905
984
1076
1172
1282
1398
1524
1658
1810
1980
2156
2350
2563
2793
3049
3316
3626
3948
4306
4695
5121
5582
0.624
0.372
0.132
0.065
0.132
0.018
0.015
0.018
0.090
0.144
0.044
0.019
0.091
0.010
0.001
0.000
0.000
0.000
0.000
0.000
0.000
2350
0.044
0.113
0.225
0.694
0.982
0.446
0.630
0.500
0.736
0.704
0.535
0.731
0.722
0.633
0.175
0.001
0.000
0.000
0.011
0.003
0.002
984
0.624
0.683
0.280
0.148
0.283
0.050
0.039
0.052
0.215
0.318
0.113
0.055
0.214
0.032
0.005
0.000
0.000
0.000
0.000
0.000
0.000
2563
0.019
0.055
0.119
0.479
0.777
0.442
0.873
0.710
0.991
0.466
0.333
0.731
0.484
0.112
0.302
0.002
0.000
0.001
0.024
0.006
0.005
1076
0.372
0.683
0.470
0.269
0.481
0.105
0.081
0.115
0.394
0.544
0.225
0.119
0.390
0.077
0.012
0.000
0.000
0.000
0.000
0.000
0.000
2793
0.091
0.214
0.390
0.947
0.734
0.914
0.415
0.324
0.484
0.985
0.794
0.722
0.484
0.860
0.090
0.000
0.000
0.000
0.005
0.001
0.001
1172
0.132
0.280
0.470
0.706
0.971
0.413
0.326
0.479
0.960
0.863
0.694
0.479
0.947
0.397
0.104
0.000
0.000
0.000
0.007
0.002
0.001
3049
0.010
0.032
0.077
0.397
0.693
0.359
0.948
0.769
0.899
0.373
0.253
0.633
0.112
0.860
0.995
0.001
0.000
0.000
0.023
0.005
0.005
1282
0.065
0.148
0.269
0.706
0.673
0.680
0.553
0.782
0.719
0.566
0.982
0.777
0.734
0.693
0.233
0.002
0.000
0.001
0.021
0.006
0.006
3316
0.001
0.005
0.012
0.104
0.233
0.087
0.416
0.582
0.289
0.082
0.050
0.175
0.302
0.090
0.995
0.036
0.003
0.014
0.174
0.670
0.062
1398
0.132
0.283
0.481
0.971
0.673
0.380
0.298
0.441
0.927
0.890
0.446
0.442
0.914
0.359
0.087
0.000
0.000
0.000
0.005
0.001
0.001
3626
0.000
0.000
0.000
0.000
0.002
0.000
0.005
0.015
0.001
0.000
0.000
0.001
0.002
0.000
0.001
0.036
0.277
0.532
0.673
0.979
0.953
1524
0.018
0.050
0.105
0.413
0.680
0.380
0.833
0.862
0.399
0.286
0.630
0.873
0.415
0.948
0.416
0.005
0.000
0.002
0.043
0.013
0.012
3948
0.000
0.000
0.000
0.000
0.000
0.000
0.000
0.001
0.000
0.000
0.000
0.000
0.000
0.000
0.000
0.003
0.277
0.713
0.192
0.371
0.386
1658
0.015
0.039
0.081
0.326
0.553
0.298
0.833
0.698
0.310
0.221
0.500
0.710
0.324
0.769
0.582
0.015
0.001
0.006
0.082
0.030
0.027
4306
0.000
0.000
0.000
0.000
0.001
0.000
0.002
0.006
0.001
0.000
0.000
0.000
0.001
0.000
0.000
0.014
0.532
0.713
0.360
0.606
0.625
1810
0.018
0.052
0.115
0.479
0.782
0.441
0.862
0.698
0.465
0.331
0.736
0.991
0.484
0.899
0.289
0.001
0.000
0.001
0.022
0.005
0.005
4695
0.000
0.000
0.000
0.007
0.021
0.005
0.043
0.082
0.022
0.004
0.002
0.011
0.024
0.005
0.023
0.174
0.673
0.192
0.360
0.691
0.670
1980
0.090
0.215
0.394
0.960
0.719
0.927
0.399
0.310
0.465
0.806
0.704
0.466
0.985
0.373
0.082
0.000
0.000
0.000
0.004
0.001
0.001
5121
0.000
0.000
0.000
0.002
0.006
0.001
0.013
0.030
0.005
0.001
0.000
0.003
0.006
0.001
0.005
0.670
0.979
0.371
0.606
0.691
0.977
Table O.7: Results from Independent Samples T-test among the frequenies
143
2156
0.144
0.318
0.544
0.863
0.566
0.890
0.286
0.221
0.331
0.806
0.535
0.333
0.794
0.253
0.050
0.000
0.000
0.000
0.002
0.000
0.000
5582
0.000
0.000
0.000
0.001
0.006
0.001
0.012
0.027
0.005
0.001
0.000
0.002
0.005
0.001
0.005
0.062
0.953
0.386
0.625
0.670
0.977
A PPENDIX P
S TATISTICAL A NALYSIS
This appendix contains a list of Tables and Figures obtained during from statistical analysis.
Table P.1 represents the HL mean and the standard desviation values of the data collected from
audiometric test.
Factor B
f(Hz)
250
500
750
1000
1500
2000
3000
4000
6000
8000
Mean (dB)
LC
0.50
4.50
4.00
2.50
5.00
4.00
1.50
6.50
11.00
-0.50
3.90
Mean HL (dB)
RC
LT
1.50
3.00
5.50
2.00
5.00
-0.50
4.50
1.50
4.00
3.00
5.00
-0.50
2.00
0.00
8.00
-2.50
13.00 6.50
-0.50 -1.00
4.85
1.15
RT
2.00
0.50
4.00
1.50
-1.00
1.00
-2.00
3.00
10.00
-3.50
1.55
Standard Deviation
LC
RC
LT
RT
4.12
4.97 6.32 8.88
3.69
3.69 6.15 6.32
7.07
4.59 5.27
5.5
6.43
3.54 4.22 3.37
4.97
4.71 6.99 5.87
4.71
5.16 4.97 6.85
4.83
4.12 5.37 7.82
8.23
5.8
6.77 7.17
5.37 10.22 5.16 8.51
6.85
6.85 6.35 8.76
Number of Subjects
Control
Target
10
10
10
10
10
10
10
10
10
10
10
10
10
10
10
10
10
10
10
10
HL Mean
(dB)
1.75
3.12
3.12
2.50
2.87
2.37
0.37
3.75
10.12
-1.37
Table P.1: Data olleted from the from audiometri test.
Table P.2 presents the DPOAE mean and the standard desviation values of the data collected
from DPOAE measurements.
145
APPENDIX P. STATISTICAL ANALYSIS
Factor B
f(Hz)
904,89
983,68
1075,53
1172,27
1281,62
1397,59
1524,34
1657,81
1809,77
1979,87
2155,95
2349,98
2562,70
2793,45
3048,78
3316,17
3626,17
3948,24
4305,73
4694,53
5120,51
5581,93
Mean (dB)
LC
-2.55
-2.41
0.21
-1.27
1.05
0.71
2.14
0.39
2.23
-0.35
0.27
0.41
2.24
1.40
0.92
2.88
5.41
6.65
7.32
5.83
5.44
4.62
4.60
Mean LFF,Aeq (dB)
RC
LT
-5.60 -0.22
-0.30 3.62
-3.56 2.27
0.17
4.22
-0.62 3.45
-1.26 2.51
0.33
4.21
1.81
5.01
2.57
3.56
1.44
2.57
1.59
1.88
2.34
2.89
2.58
2.94
1.46
1.76
2.02
3.72
4.53
4.48
6.39
7.88
7.59
9.30
7.97
8.57
4.89
7.77
6.08
9.57
5.67
9.24
4.27
1.97
RT
3.64
1.70
2.43
2.54
4.31
3.30
3.99
5.69
2.96
2.32
2.62
3.36
3.62
2.34
3.54
3.73
5.93
7.43
6.67
6.84
7.16
7.94
2.18
LC
6.93
6.39
5.52
8.23
6.72
6.53
6.63
9.27
6.38
7.48
6.16
5.53
4.87
5.89
5.43
6.65
5.86
6.42
6.95
7.44
8.44
8.54
Standard Deviation
RC
LT
RT
4.90 6.37 6.81
5.37 4.88 4.45
5.34 6.59 4.78
6.22 6.64 6.10
8.40 6.71 7.02
5.50 8.51 5.45
5.23 7.30 7.11
5.63 6.19 5.54
3.79 5.10 4.96
4.30 4.96 6.62
4.95 4.57 4.21
5.78 4.82 6.05
4.32 6.21 6.31
3.64 7.65 6.07
4.38 5.08 5.86
5.18 6.47 6.15
6.05 5.79 4.27
6.26 5.40 5.36
8.24 4.85 4.86
10.32 5.17 4.67
8.42 3.84 3.57
9.29 3.61 3.90
Number of Subjects
Control
Target
10
10
10
10
10
10
10
10
10
10
10
10
10
10
10
10
10
10
10
10
10
10
10
10
10
10
10
10
10
10
10
10
10
10
10
10
10
10
10
10
10
10
10
10
Table P.2: Data olleted from DPOAE measurements.
146
DPOAE Mean
(dB)
-1.18
0.65
0.33
1.41
2.04
1.31
2.66
3.22
2.83
1.49
1.59
2.25
2.84
1.74
2.55
3.90
6.40
7.74
7.63
6.33
7.06
6.86
Was this manual useful for you? yes no
Thank you for your participation!

* Your assessment is very important for improving the work of artificial intelligence, which forms the content of this project

Download PDF

advertisement