142 CHAPTER 4 REVERBERATION 4.1 GENERAL Understanding speech is hindered by the combined effects of excessive noise and reverberation in the classroom, which tends to interfere in the learning process. The combination of noise and reverberation exerts a stronger negative effect on speech recognition than the sum of their separate effects (Bradley et al 1999, Crandell and Smaldino 2000). Having discussed the background noise in the previous chapter, the effects of reverberation in the classrooms would be discussed in this chapter. An acoustical phenomenon that occurs in an enclosed space, such as a classroom, when sound waves persist or prolong in that space as a result of repeated reflections or scattering from hard surfaces enclosing the space or objects in the space, such as chairs or cabinets is called reverberation (Siebein 1994, Siebein et al 1997). An important variable that determines the acoustics of the classroom is reverberation. If there is a high degree of reverberation, the sound reaching a listener will be subjected to a prolonged number of reflections. Direct sound will arrive at the listener first, followed by many reflections of the sound. For a space designed for speech, late reflections can destroy Speech Intelligibility as these reflections of earlier syllables of the speech mask subsequent syllables, and thus the intelligibility of the speech is degraded. By contrast, reflections which arrive at a listener soon after the direct sound strengthen the direct signal and enhance intelligibility. Early (arriving up to 35 ms after the direct sound) sound waves reaching the children from the teacher enhance the
38
Embed
CHAPTER 4 REVERBERATION - Shodhgangashodhganga.inflibnet.ac.in/bitstream/10603/9572/33/12...142 CHAPTER 4 REVERBERATION 4.1 GENERAL Understanding speech is hindered by the combined
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
142
CHAPTER 4
REVERBERATION
4.1 GENERAL
Understanding speech is hindered by the combined effects of
excessive noise and reverberation in the classroom, which tends to interfere in
the learning process. The combination of noise and reverberation exerts a
stronger negative effect on speech recognition than the sum of their separate
effects (Bradley et al 1999, Crandell and Smaldino 2000). Having discussed
the background noise in the previous chapter, the effects of reverberation in
the classrooms would be discussed in this chapter. An acoustical phenomenon
that occurs in an enclosed space, such as a classroom, when sound waves
persist or prolong in that space as a result of repeated reflections or scattering
from hard surfaces enclosing the space or objects in the space, such as chairs
or cabinets is called reverberation (Siebein 1994, Siebein et al 1997). An
important variable that determines the acoustics of the classroom is
reverberation. If there is a high degree of reverberation, the sound reaching a
listener will be subjected to a prolonged number of reflections. Direct sound
will arrive at the listener first, followed by many reflections of the sound. For
a space designed for speech, late reflections can destroy Speech Intelligibility
as these reflections of earlier syllables of the speech mask subsequent
syllables, and thus the intelligibility of the speech is degraded. By contrast,
reflections which arrive at a listener soon after the direct sound strengthen the
direct signal and enhance intelligibility. Early (arriving up to 35 ms after the
direct sound) sound waves reaching the children from the teacher enhance the
143
hearing quality for children (Whitlock and Dodd 2008). It was also reported
by him that short Reverberation Times may actually be beneficial to the
listener if the speech signal is of insufficient intensity.
4.2 REVERBERATION TIME
The traditional parameter which has been studied in evaluating the
effect of room acoustics is Reverberation Time (RT). Reverberation Time is
defined as the time (in seconds) it takes for the sound from a source to
decrease in level by 60 dB (Sabine 1964) after the source has stopped. A
decrease of 60 dB represents a reduction of 1/1,000,000 of the original
intensity of the sound. A formula to calculate RT was described in 1964 by
Sabine and famously referred to as Sabine formula (Sabine 1964):
60
0.161VRT
S( )= S a (4.1)
where RT60 = RT in seconds (The suffix 60 indicates a decay of sound 60
dB), (0.161 is a constant if room volume is stated in meter3), V = room
volume in m3 and S = the surface area in m
2 of the various materials in the
room including the students absorption area; α = respective absorption
coefficients at a given frequency. From the RT formula described above, it
can be seen that there are two basic factors that affect the RT in a room. The
first is the room volume. The larger the room volume, the longer the RT will
be. The second variable is the amount of sound absorption in the room. The
greater the area of such materials, the shorter the RT. Room reverberation
varies as a function of frequency and, therefore, may need to be measured at
discrete frequencies. RT is often reported as the mean decay time at 500, 1000
and 2000 Hz. This average describes the characteristics of most rooms fairly
well, because most materials do not absorb low frequencies well, room
reverberation is shorter at higher frequencies and longer in lower frequency
144
regions. It is recommended that RT be measured at discrete frequencies from
125 to 8000 Hz, whenever excessive reverberation seems to interfere with
communication. Such information could significantly aid the audiologist in
determining the appropriate degree and type of absorptive materials needed
for a reduction of RT in that environment.
The Early Decay Time (EDT) parameter is the Reverberation Time,
measured over the decay of the first 10 dB. The EDT is also used instead of
RT in Speech Intelligibility predictions and it can be represented as RT10.
RT20 is the Reverberation Time of the room evaluated over a 20 dB decay
range (from -5 to -25 dB). It is the time distance between the -5 dB and the -
25 dB. RT30 is the reverberation time of the room evaluated over a 30 dB
decay range (from -5 to -35 dB). RT 20 and RT 30 can be evaluated from the
decay plot using regression analysis. The decay plot denotes the reduction in
sound level from 0 to - 60 and hence the decay ranges are shown as negative.
The RT for classrooms are mostly stipulated by standards in the
unoccupied conditions of classes as seen in Table 2.7, Chapter 2 and it
depends on the size of the classrooms, as RT is directly proportional to the
volume of the enclosures and inversely proportional to the absorbing areas.
The effects of reverberation can easily be felt in a completely empty room
when there is nothing to absorb the sound. Such rooms have long RT’s and
are often described as ‘echoy’ or ‘hollow’. The addition of appropriate
amount of absorbing material into a room removes this effect and lowers RT.
WHO (1999) and ANSI S12.6 (2002) recommend the RT of school
classrooms to be 0.6 s for classrooms having volumes less than 283 m3 and,
for volumes up to double this value, up to 0.7 s. The RT is for the unoccupied
condition of the class in most of the regulations. In the UK the Building
Bulletin (2003) has proposed RT less than 0.6 s for primary school
classrooms and, for secondary schools, up to 0.8 s, but for the hearing
145
impaired a lower RT of 0.4s is suggested. The American Speech Language
Hearing Association (1995) has proposed a value of less than 0.4 s and the
British Association of Teachers of the Deaf (BATOD 2001) has suggested a
RT of less than 0.4 s. The (AS/NZ 2000), the New Zealand standards suggest
RT of 0.4 s - 0.5 s for primary school teaching spaces. The Indian National
Building Code (NBC 2005) proposed a value of 0.75 s for an occupied class
and a higher value of 1.25 s for the empty classroom.
4.3 MEASUREMENT OF REVERBERATION TIME
A two channel Bruel and Kjaer (B&K) 2250 as seen in Figures 4.1a
and b, modular real time sound analyzer has been used to measure
Reverberation Time inside occupied and empty classrooms. Figure 4.1a
shows the sound level metre used in the hand held position without a tri-pod
in occupied class as the use of a tri-pod disturbs the normal class functioning.
Figure 4.1b shows the sound level metre placed on the tri-pod used for
measuring unoccupied class. It automatically calculates the mean
Reverberation Time for each frequency. The measurements were taken
following the specifications of the ISO 3382 (1997). RT measurements were
taken at three different points R1, R2 and R3 in the classroom as shown in
Figure 4.2. Three readings were taken at each point (Figures 4.3, 4.4 and 4.5)
and the mean was calculated.
The measurements were then transferred to a computer using
Qualifier 7830 software from Bruel and Kjaer, which calculated the mean
Reverberation Time for each frequency. This procedure was repeated for all
the classrooms in which RT was measured. The impact sound signal was
produced by clapping of hands and the decay was recorded which gave the
Reverberation Time.
146
(a) (b)
Figure 4.1 (a) Bruel and Kjaer (BK 2250) equipment used for
measuring Reverberation Time (b) BK 2250 placed on a tri-
pod to measure RT in unoccupied classroom
Measured data have been compared with reference values found in
the standards ANS1 S12.60 (2002) and DIN 18041 (2004), NBC (2005). The
classrooms measured had the following internal finishes--floor and ceiling
were concrete, finished with cement plaster and painted. The four walls were
of brick with cement plaster and painted. The seats in the classroom were
mostly wood painted or varnished with some classrooms having metal painted
or plastic seats. Doors and windows provided were kept open and so the
absorption coefficients corresponding to the openings were provided.
147
* *
R3 R1
*R2
Figure 4.2 RT measured in three positions R1, R2 and R3 in both
occupied and unoccupied classrooms
Figure 4.3 Right Side Position of RT measurement at R1
Figure 4.4 RT measured in the centre position R2 of the classroom
148
RT was measured using BK 2250 by producing a sufficiently loud
impulse noise and as the sound dies away the trace on the level recorder will
show a distinct slope. Analysis of this slope reveals the measured
Reverberation Time. As BK 2250 is a modern digital sound level meter it can
carry out this analysis automatically. The loud impulse noise was created by
clapping of hands and a few seconds were given to record the trace of the
sound dying. The recorded value gave the RT in seconds for various frequencies.
Figure 4.5 Left side position of RT measurement at R3
4.3.1 Reverberation Time Measured in Unoccupied class
Reverberation Time was measured in furnished unoccupied
classrooms which had a volume and a seat capacity as ranging in the
Table 3.1, 3.2 and 3.3 of Chapter 3, for the three sites of Noisy, Housing and
Quiet. RT is a function of the volume and the internal characteristics of the
room and is not influenced by the zone in which the school is located, so the
discussion of RT is made in total and not as per the zones. In unoccupied
149
classroom the absorption is by the room interior surfaces and by the furniture
present in the class. The absorption of sound by the children is absent and so
the RT for unoccupied class would be more than the RT for occupied class.
4.3.2 Reverberation Time Measured in Occupied Class
When the students have occupied the class and class was ongoing
the RT was measured. The students were made to sit quietly so that the
impact sound created was heard clearly. The RT in occupied condition would
be less than that in unoccupied class due to student absorption. Table 4.1
shows the unoccupied and occupied values of RT measured using BK 2250
and as a sample, only 30 classrooms measured are shown.