I
AE
S In
t
er
na
t
io
na
l J
o
urna
l o
f
Art
if
icia
l In
t
ellig
ence
(
I
J
-
AI)
Vo
l.
1
3
,
No
.
4
,
Dec
em
b
er
2
0
2
4
,
p
p
.
4
9
9
8
~
5
0
0
4
I
SS
N:
2
2
5
2
-
8
9
3
8
,
DOI
: 1
0
.
1
1
5
9
1
/ijai.v
1
3
.
i
4
.
p
p
4
9
9
8
-
5
0
0
4
4998
J
o
ur
na
l ho
m
ep
a
g
e
:
h
ttp
:
//ij
a
i
.
ia
esco
r
e.
co
m
Spectral spli
tt
ing
o
f
speech
sig
na
l using
t
ime v
a
ry
ing
recursiv
e
filt
ers
for bina
ura
l hearing
aids
Apa
rna
Chila
k
a
wa
d,
P
a
nd
ura
ng
a
ra
o
N.
K
ulk
a
rni
D
e
p
a
r
t
me
n
t
o
f
El
e
c
t
r
o
n
i
c
s
a
n
d
C
o
mm
u
n
i
c
a
t
i
o
n
En
g
i
n
e
e
r
i
n
g
,
B
a
sa
v
e
sh
w
a
r
En
g
i
n
e
e
r
i
n
g
C
o
l
l
e
g
e
(
a
f
f
i
l
i
a
t
e
d
t
o
V
i
s
v
e
s
v
a
r
a
y
a
Te
c
h
n
o
l
o
g
i
c
a
l
U
n
i
v
e
r
si
t
y
)
,
B
a
g
a
l
k
o
t
e
,
I
n
d
i
a
Art
icle
I
nfo
AB
S
T
RAC
T
A
r
ticle
his
to
r
y:
R
ec
eiv
ed
No
v
2
2
,
2
0
2
3
R
ev
is
ed
Ma
r
2
3
,
2
0
2
4
Acc
ep
ted
Ap
r
1
7
,
2
0
2
4
S
p
e
e
c
h
p
e
rc
e
p
ti
o
n
i
n
n
o
isy
e
n
v
ir
o
n
m
e
n
ts
is
re
d
u
c
e
d
in
h
u
m
a
n
wit
h
se
n
so
rin
e
u
ra
l
h
e
a
rin
g
lo
ss
(S
NH
L)
d
u
e
t
o
m
a
sk
in
g
.
M
o
d
e
ra
te
S
N
HL
c
a
n
n
o
t
b
e
c
u
re
d
m
e
d
ica
ll
y
h
e
n
c
e
m
a
sk
in
g
e
ffe
c
ts
sh
o
u
l
d
b
e
re
d
u
c
e
d
t
o
e
n
h
a
n
c
e
sp
e
e
c
h
p
e
rc
e
p
ti
o
n
.
T
o
re
d
u
c
e
m
a
sk
in
g
,
p
ro
c
e
ss
in
g
d
e
lay
a
n
d
h
a
rd
wa
re
c
o
m
p
lex
it
y
th
e
p
re
se
n
t
p
a
p
e
r
is
p
ro
p
o
se
d
a
sc
h
e
m
e
to
p
a
rti
ti
o
n
th
e
v
o
ice
sig
n
a
l
in
t
o
two
sig
n
a
ls
w
h
ich
a
re
c
o
m
p
lem
e
n
tary
t
o
e
a
c
h
o
t
h
e
r
b
y
u
si
n
g
t
h
e
fil
ter
-
b
a
n
k
s
u
m
m
a
ti
o
n
m
e
th
o
d
(F
BS
M
)
with
a
se
t
o
f
ti
m
e
-
v
a
ry
i
n
g
re
c
u
rsiv
e
b
a
n
d
p
a
ss
fil
ters
.
P
e
rfo
rm
a
n
c
e
o
f
th
e
fil
ter
is
e
v
a
lu
a
ted
with
fo
ll
o
wi
n
g
m
e
a
su
re
s:
p
e
rc
e
p
tu
a
l
e
v
a
lu
a
ti
o
n
o
f
sp
e
e
c
h
q
u
a
li
t
y
(P
ES
Q),
m
e
a
n
o
p
in
i
o
n
sc
o
re
(M
OS)
fo
r
sp
e
e
c
h
q
u
a
li
ty
a
n
d
m
o
d
ifi
e
d
r
h
y
m
e
tes
t
(M
R
T)
fo
r
s
p
e
e
c
h
in
telli
g
ib
il
it
y
.
Th
e
tes
t
sig
n
a
ls
u
se
d
fo
r
th
e
e
v
a
l
u
a
ti
o
n
o
f
q
u
a
li
ty
a
re
a
sy
ll
a
b
le
a
n
d
a
wo
r
d
a
n
d
f
o
r
t
h
e
e
v
a
lu
a
ti
o
n
o
f
i
n
telli
g
ib
il
it
y
3
0
0
m
o
n
o
sy
l
lab
ic
wo
rd
s
a
re
u
se
d
.
Th
e
re
su
lt
s
d
e
m
o
n
stra
ted
a
n
in
c
re
a
se
in
th
e
q
u
a
li
ty
a
n
d
in
telli
g
ib
il
it
y
o
f
p
r
o
c
e
ss
e
d
sp
e
e
c
h
i
n
a
n
o
isy
e
n
v
ir
o
n
m
e
n
t.
As
a
r
e
su
lt
,
t
h
e
re
is an
e
n
h
a
n
c
e
m
e
n
t
in
p
e
rc
e
p
ti
o
n
o
f
p
ro
c
e
ss
e
d
sp
e
e
c
h
i
n
a
n
o
isy
e
n
v
i
ro
n
m
e
n
t
.
K
ey
w
o
r
d
s
:
Ma
s
k
in
g
Mo
d
if
ied
r
h
y
m
e
test
Sen
s
o
r
in
eu
r
al
h
ea
r
in
g
lo
s
s
Sp
ee
ch
q
u
ality
T
im
e
-
v
ar
y
in
g
r
ec
u
r
s
iv
e
f
ilter
s
T
h
is i
s
a
n
o
p
e
n
a
c
c
e
ss
a
rticle
u
n
d
e
r th
e
CC B
Y
-
SA
li
c
e
n
se
.
C
o
r
r
e
s
p
o
nd
ing
A
uth
o
r
:
Ap
ar
n
a
C
h
ilak
awa
d
Dep
ar
tm
en
t o
f
E
lectr
o
n
ics an
d
C
o
m
m
u
n
icatio
n
E
n
g
in
ee
r
i
n
g
,
B
asav
esh
war
E
n
g
in
ee
r
in
g
C
o
lleg
e
af
f
iliated
to
Vis
v
esv
ar
ay
a
T
ec
h
n
o
lo
g
ical
U
n
iv
er
s
ity
B
ag
alk
o
te
,
B
elag
av
i,
I
n
d
ia
E
m
ail:
ap
ar
n
a
c
_
1
2
@
r
ed
if
f
m
ai
l.c
o
m
1.
I
NT
RO
D
UCT
I
O
N
T
h
er
e
ar
e
d
if
f
er
e
n
t
k
in
d
s
o
f
h
e
ar
in
g
lo
s
s
;
s
en
s
o
r
in
eu
r
al
h
ea
r
in
g
lo
s
s
(
SNHL
)
co
n
d
u
ctiv
e
h
e
ar
in
g
lo
s
s
an
d
m
ix
ed
h
ea
r
i
n
g
lo
s
s
[
1
]
.
SNHL
ar
is
es
o
w
in
g
to
th
e
b
r
o
a
d
en
in
g
o
f
au
d
ito
r
y
f
ilter
s
o
f
th
e
co
ch
lea
in
ev
itab
ly
lead
in
g
to
th
e
m
ask
in
g
o
f
f
r
eq
u
en
cy
c
o
m
p
o
n
en
ts
o
f
ad
ja
ce
n
t
au
d
ito
r
y
f
ilter
b
a
n
d
s
.
A
s
a
r
esu
lt,
s
p
ee
ch
p
er
ce
p
tio
n
d
im
in
is
h
es
in
h
u
m
an
with
SNHL
.
Du
e
to
th
is
,
th
e
s
en
s
o
r
y
s
y
s
tem
'
s
ca
p
ac
ity
o
f
f
r
eq
u
en
c
y
r
eso
lu
tio
n
d
ec
r
ea
s
es
an
d
s
p
ee
c
h
d
etec
tio
n
in
n
o
is
y
e
n
v
ir
o
n
m
en
ts
b
ec
o
m
es
ch
allen
g
in
g
[
2
]
,
[
3
]
.
E
a
r
lier
s
tu
d
ies
d
em
o
n
s
tr
ated
th
at
th
e
u
s
e
o
f
a
p
air
o
f
co
m
b
f
ilter
s
an
d
tim
e
-
v
ar
y
i
n
g
co
m
b
f
ilter
s
(
f
in
i
te
im
p
u
ls
e
r
esp
o
n
s
e
(
FIR)
)
to
p
ar
titi
o
n
th
e
v
o
ice
s
ig
n
al
in
to
two
co
m
p
lem
en
tar
y
s
p
ec
tr
a
f
o
r
b
in
au
r
al
h
ea
r
i
n
g
a
id
was
ef
f
ec
tiv
e
in
d
ec
r
ea
s
in
g
th
e
im
p
ac
t
o
f
f
r
e
q
u
en
cy
m
ask
in
g
[
4
]
–
[
8
]
.
Alth
o
u
g
h
it
was
d
is
co
v
er
e
d
th
at
th
e
ea
r
lier
m
eth
o
d
s
wer
e
s
u
cc
ess
f
u
l
in
tack
lin
g
f
r
eq
u
en
cy
m
ask
in
g
,
th
e
s
tep
alter
in
th
e
m
ag
n
itu
d
e
r
esp
o
n
s
es
w
ith
a
f
ix
ed
am
o
u
n
t
o
f
s
ep
ar
ate
tim
e
s
h
if
ts
af
f
ec
ts
th
e
ab
ilit
y
to
id
en
tify
g
ap
s
.
Fu
r
th
er
,
s
in
ce
FIR
f
ilter
s
u
s
ed
in
th
e
ea
r
lier
s
tu
d
ies
g
en
er
ally
r
eq
u
ir
e
h
ig
h
er
o
r
d
er
,
th
er
eb
y
in
cr
ea
s
in
g
th
e
p
r
o
ce
s
s
in
g
d
elay
a
n
d
h
ar
d
war
e
co
m
p
lex
ity
th
u
s
m
a
k
in
g
it
d
if
f
icu
lt
f
o
r
r
ea
l
-
tim
e
im
p
l
em
en
tatio
n
.
Hen
ce
,
we
p
r
o
p
o
s
ed
th
e
im
p
lem
en
tatio
n
o
f
th
e
s
p
ec
tr
al
s
p
litt
in
g
u
s
in
g
th
e
f
ilter
-
b
an
k
s
u
m
m
atio
n
m
eth
o
d
(
FB
SM)
with
a
s
et
o
f
tim
e
-
v
a
r
y
in
g
in
f
in
ite
im
p
u
ls
e
r
esp
o
n
s
e
(
I
I
R
)
b
an
d
-
p
ass
f
ilter
s
f
o
r
b
in
au
r
al
h
ea
r
in
g
aid
s
,
wh
ich
g
en
er
ally
r
eq
u
ir
e
m
u
c
h
less
o
r
d
er
an
d
h
ar
d
war
e
co
m
p
lex
ity
[
9
]
.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
n
t J Ar
tif
I
n
tell
I
SS
N:
2
2
5
2
-
8
9
3
8
S
p
ec
tr
a
l sp
litt
in
g
o
f sp
ee
ch
s
ig
n
a
l u
s
in
g
time
va
r
yin
g
r
ec
u
r
s
i
ve
filt
ers
fo
r
b
in
a
u
r
a
l …
(
A
p
a
r
n
a
C
h
ila
ka
w
a
d
)
4999
T
h
e
o
b
jectiv
e
o
f
t
h
is
p
ap
e
r
is
to
r
ed
u
ce
t
h
e
h
a
r
d
war
e
c
o
m
p
lex
ity
o
f
b
in
a
u
r
al
h
ea
r
in
g
aid
s
a
n
d
im
p
r
o
v
e
s
p
ee
ch
p
er
ce
p
tio
n
in
SNHL
p
er
s
o
n
s
b
y
r
ed
u
cin
g
m
ask
in
g
as
th
is
l
o
s
s
ca
n
n
o
t
b
e
m
ed
ically
cu
r
ed
.
T
h
is
is
ac
h
iev
ed
u
s
in
g
a
b
a
n
k
o
f
tim
e
-
v
ar
y
i
n
g
I
I
R
b
a
n
d
-
p
ass
f
ilter
s
to
p
ar
titi
o
n
v
o
ic
e
s
ig
n
al
in
to
two
co
m
p
lem
en
tar
y
s
p
ec
tr
a
f
o
r
b
in
au
r
al
d
ic
h
o
tic
h
ea
r
i
n
g
ai
d
.
Usi
n
g
m
o
d
if
ied
r
h
y
m
e
te
s
t
(
MRT)
[
1
0
]
f
o
r
in
tellig
ib
ilit
y
an
d
p
er
ce
p
tu
al
ev
alu
atio
n
o
f
s
p
ee
ch
q
u
ality
(
PESQ)
[
1
1
]
,
an
d
m
ea
n
o
p
i
n
io
n
s
co
r
e
(
MO
S)
[
1
2
]
–
[
1
4
]
f
o
r
q
u
ality
th
e
ef
f
icien
cy
o
f
th
e
ap
p
r
o
ac
h
in
en
h
an
cin
g
s
p
ee
ch
p
er
ce
p
tio
n
was
as
s
es
s
ed
.
T
o
m
in
im
ize
f
r
eq
u
en
cy
m
ask
in
g
,
it
h
as
b
ee
n
p
r
o
p
o
s
ed
t
o
u
s
e
a
s
et
o
f
tim
e
-
v
a
r
y
in
g
b
a
n
d
-
p
a
s
s
(
I
I
R
)
f
ilter
s
with
FB
SM
th
at
s
wee
p
s
co
n
tin
u
o
u
s
ly
in
m
ag
n
itu
d
e
r
esp
o
n
s
es,
wh
ich
u
ltima
tely
en
h
a
n
ce
s
s
p
ee
ch
p
er
ce
p
tio
n
.
Acc
o
r
d
in
g
to
San
g
ee
th
a
a
n
d
Kan
n
an
[
1
5
]
,
v
a
r
io
u
s
d
ig
ital
f
ilter
s
ar
e
d
is
cu
s
s
ed
.
T
h
e
in
p
u
t
s
ig
n
al
is
th
e
v
o
ice
s
ig
n
al.
W
ith
th
e
i
n
p
u
t
v
o
ice
s
ig
n
al,
a
d
d
itiv
e
wh
ite
Gau
s
s
ian
n
o
is
e
(
AW
GN)
is
ad
d
ed
.
Usi
n
g
s
am
p
lin
g
r
ate
co
n
v
er
s
io
n
,
th
e
n
o
is
y
s
p
ee
ch
s
ig
n
al
s
p
ec
tr
u
m
is
d
o
wn
-
s
am
p
led
in
to
m
an
y
s
am
p
lin
g
r
ates.
T
h
e
n
o
is
y
au
d
io
s
ig
n
al
an
d
its
s
u
b
-
b
a
n
d
s
ar
e
s
u
b
jecte
d
to
a
n
u
m
b
er
o
f
tr
a
n
s
f
o
r
m
s
i
n
clu
d
in
g
th
e
f
ast
f
o
u
r
ie
r
tr
an
s
f
o
r
m
(
FF
T
)
,
f
ast
walsh
h
ad
am
ar
d
t
r
an
s
f
o
r
m
(
FW
HT
)
,
an
d
d
is
cr
ete
wav
elet
tr
an
s
f
o
r
m
(
DW
T
)
.
T
h
e
r
ec
tan
g
u
lar
,
h
a
n
n
in
g
,
h
am
m
in
g
,
b
lac
k
m
an
,
an
d
k
aiser
win
d
o
ws
ar
e
u
s
ed
in
th
e
d
esig
n
o
f
FIR
f
ilter
s
,
wh
er
ea
s
b
u
tter
wo
r
th
an
d
ch
e
b
y
s
h
ev
f
ilter
s
ar
e
u
s
ed
i
n
th
e
d
esig
n
an
d
im
p
lem
en
tatio
n
o
f
I
I
R
f
ilter
s
.
T
h
e
f
ilter
co
ef
f
icien
ts
ar
e
th
e
n
s
u
b
jecte
d
to
q
u
an
tizatio
n
.
L
astl
y
,
th
e
s
ig
n
al
to
q
u
an
tizatio
n
r
atio
(
SQNR
)
is
u
s
ed
to
ev
alu
ate
th
e
p
er
f
o
r
m
an
ce
o
f
th
e
f
ilter
s
.
Acc
o
r
d
in
g
to
th
e
p
er
f
o
r
m
an
ce
m
ea
s
u
r
es,
DW
T
ap
p
lied
s
ig
n
als f
ilter
ed
b
y
th
e
ch
eb
y
s
h
ev
h
ig
h
p
ass
f
ilter
(
HPF)
g
iv
e
h
ig
h
SQNR
i
n
I
I
R
f
ilter
in
g
,
a
n
d
DW
T
ap
p
lied
s
ig
n
als
f
ilter
ed
b
y
th
e
b
lack
m
an
win
d
o
w
p
r
o
v
id
e
h
ig
h
SQNR
in
FIR
f
ilte
r
in
g
.
T
h
e
ass
ess
m
en
t
o
f
th
e
p
er
f
o
r
m
an
ce
o
f
th
e
f
ilter
co
ef
f
icien
ts
is
b
ased
o
n
t
h
e
SQNR
.
T
h
e
r
esear
ch
er
s
in
[
1
6
]
,
[
1
7
]
co
m
p
ar
ed
o
f
b
i
n
au
r
al
h
ea
r
i
n
g
aid
with
th
e
m
o
n
au
r
al
h
ea
r
in
g
aid
is
d
is
cu
s
s
ed
.
Am
in
i
et
a
l.
[
1
8
]
p
r
o
p
o
s
ed
a
n
o
is
e
r
ed
u
ctio
n
m
eth
o
d
th
at
p
r
eser
v
e
d
th
e
b
in
a
u
r
al
s
p
atial
cu
es
o
f
p
o
in
t
s
o
u
r
ce
s
an
d
s
im
u
ltan
e
o
u
s
ly
d
eter
m
in
ed
th
e
b
est
r
ate
allo
ca
tio
n
an
d
th
e
o
p
tim
al
e
s
tim
atio
n
weig
h
ts
ac
r
o
s
s
all
s
en
s
o
r
s
an
d
f
r
eq
u
en
cies.
T
o
r
etain
th
e
b
in
au
r
al
s
p
atial
cu
es
o
f
p
o
in
t
s
o
u
r
ce
s
,
th
e
s
u
g
g
ested
tech
n
iq
u
e
tak
es
in
to
ac
co
u
n
t
b
o
th
lin
ea
r
an
d
r
ate
c
o
n
s
tr
ain
ts
.
T
h
e
in
ter
-
a
u
r
al
lev
el
d
if
f
er
en
ce
(
I
L
D)
er
r
o
r
,
t
h
e
in
ter
-
au
r
al
tim
e
d
if
f
er
en
ce
(
I
T
D)
er
r
o
r
,
an
d
th
e
av
er
a
g
ed
b
i
n
au
r
al
SNR
wer
e
u
s
ed
t
o
ass
ess
th
e
ef
f
icien
cy
o
f
th
e
s
u
g
g
ested
ap
p
r
o
ac
h
.
T
h
e
r
esu
lts
d
em
o
n
s
tr
ated
th
at
th
e
s
u
g
g
ested
ap
p
r
o
ac
h
p
er
f
o
r
m
ed
b
etter
th
an
s
p
atially
ac
cu
r
ate
n
o
is
e
r
ed
u
ctio
n
tech
n
iq
u
es
u
tili
zin
g
r
an
d
o
m
r
ate
allo
ca
tio
n
s
ch
em
es.
T
h
e
d
e
v
e
lo
p
m
en
t
o
f
s
ev
er
al
ef
f
icien
t
s
p
ee
ch
p
r
o
ce
s
s
in
g
al
g
o
r
ith
m
s
f
o
r
p
o
r
tab
le
d
ev
ices
is
f
ac
ilit
ated
b
y
t
h
e
c
u
r
r
en
t
t
r
en
d
o
f
lo
w
p
o
wer
an
d
ar
ea
co
n
s
u
m
p
tio
n
[
1
9
]
.
Hi
g
h
p
er
f
o
r
m
a
n
ce
is
n
o
t
th
e
m
ai
n
ar
ea
o
f
em
p
h
asis
in
b
io
lo
g
ic
al
ap
p
licatio
n
s
lik
e
h
ea
r
in
g
aid
s
.
Po
wer
an
d
v
icin
i
ty
ar
e
eq
u
ally
im
p
o
r
tan
t f
ac
to
r
s
.
T
h
e
m
ar
k
et
p
en
etr
atio
n
is
af
f
ec
ted
b
y
a
v
ar
iety
o
f
th
in
g
s
.
T
h
e
s
y
s
tem
ca
n
th
u
s
o
p
er
ate
as
s
lo
wly
as
p
o
s
s
ib
le
to
tr
ad
e
s
p
ee
d
f
o
r
p
o
we
r
an
d
v
icin
ity
.
A
n
ef
f
ec
tiv
e
f
o
ld
e
d
I
I
R
f
ilter
is
d
ev
elo
p
ed
.
T
h
e
f
o
ld
ed
d
esig
n
e
m
p
lo
y
s
th
e
id
ea
o
f
tem
p
o
r
al
m
u
ltip
lex
in
g
,
a
n
d
all
o
p
er
atio
n
s
ar
e
h
an
d
led
b
y
ju
s
t
o
n
e
m
u
ltip
lier
an
d
o
n
e
ad
d
er
.
I
ts
ef
f
ec
tiv
en
ess
is
ev
alu
ated
in
co
m
p
ar
is
o
n
with
a
ty
p
ical
I
I
R
f
ilter
im
p
lem
en
te
d
in
d
ir
ec
t f
o
r
m
-
2
s
tr
u
ctu
r
e.
I
n
v
iv
ad
o
h
ig
h
-
lev
el
s
y
n
th
esis
(
HL
S)
a
FIR f
ilter
is
im
p
lem
en
ted
an
d
in
s
tan
tiated
in
th
e
b
lo
ck
d
esig
n
.
T
h
e
lo
o
k
u
p
tab
le
(
L
UT
)
u
tili
za
tio
n
is
r
ed
u
ce
d
f
r
o
m
0
.
1
3
to
0
.
0
9
.
T
h
e
L
UT
u
tili
za
tio
n
in
I
O
an
d
g
lo
b
al
clo
ck
b
u
f
f
er
(
B
UFG
)
r
em
ain
s
th
e
s
am
e.
2.
M
E
T
H
O
DO
L
O
G
Y
I
n
co
ch
lea
s
p
ec
tr
al
co
m
p
o
n
e
n
ts
o
f
n
ea
r
b
y
b
a
n
d
s
ar
e
lik
ely
t
o
o
v
er
la
p
lead
in
g
to
f
r
eq
u
e
n
c
y
m
ask
in
g
.
Fo
r
th
is
r
ea
s
o
n
,
o
d
d
b
an
d
s
ar
e
p
r
esen
ted
to
th
e
lef
t
ea
r
an
d
ev
en
b
an
d
s
to
th
e
r
ig
h
t
ea
r
to
less
en
m
ask
in
g
.
T
o
p
r
esen
t
s
p
ee
ch
s
ig
n
al
d
ich
o
ti
ca
lly
(
to
th
e
lef
t
an
d
r
i
g
h
t
ea
r
s
)
with
co
m
p
le
m
en
tar
y
s
p
ec
tr
al
co
m
p
o
n
en
ts
,
a
FB
SM
with
a
s
et
(
b
an
k
o
f
2
2
)
o
f
tim
e
-
v
ar
y
in
g
b
an
d
-
p
ass
(
r
ec
u
r
s
iv
e)
f
ilter
s
o
f
o
r
d
er
2
5
w
ith
co
m
p
lem
e
n
tar
y
m
ag
n
itu
d
e
r
esp
o
n
s
es
is
u
s
ed
.
I
n
th
is
tech
n
iq
u
e,
s
p
ee
ch
s
i
g
n
al
is
p
r
o
ce
s
s
ed
with
b
an
k
o
f
b
a
n
d
-
p
ass
f
ilter
s
(
with
d
if
f
er
e
n
t
b
a
n
d
wid
th
s
alt
er
n
ately
s
u
m
m
e
d
)
to
p
r
esen
t
s
ig
n
al
to
SNHL
lis
ten
er
s
.
T
h
e
m
ag
n
itu
d
e
r
esp
o
n
s
e
|H
m
(
f
)
|
an
d
im
p
u
ls
e
r
esp
o
n
s
e
h
m
(
n
)
o
f
a
tim
e
-
v
ar
y
in
g
b
an
d
-
p
ass
f
ilter
s
ar
e
d
y
n
am
ic
an
d
a
r
e
f
u
n
ctio
n
s
o
f
tim
e
'
m
'
.
T
h
e
p
r
o
p
o
s
ed
ap
p
r
o
ac
h
e
m
p
lo
y
s
a
s
et
o
f
tim
e
-
v
ar
y
i
n
g
b
an
d
-
p
ass
I
I
R
f
ilter
s
with
m
ag
n
itu
d
e
r
esp
o
n
s
es
th
at
ar
e
s
wep
t
alo
n
g
th
e
f
r
eq
u
en
cy
ax
is
with
r
esp
ec
t
to
tim
e
to
p
ar
titi
o
n
th
e
s
p
ee
ch
in
to
two
s
ig
n
als
wit
h
co
m
p
lem
en
tar
y
s
p
ec
tr
a
f
o
r
b
i
n
au
r
al
d
ich
o
tic
h
ea
r
in
g
aid
.
F
ig
u
r
e
1
s
h
o
ws
a
b
lo
c
k
s
ch
em
atic
o
f
t
h
e
p
r
o
p
o
s
ed
wo
r
k
'
s
m
eth
o
d
o
lo
g
y
.
T
h
e
v
o
i
ce
s
ig
n
al
is
d
iv
id
ed
in
to
two
p
o
r
tio
n
s
with
co
m
p
lem
e
n
tar
y
s
p
ec
tr
al
co
m
p
o
n
en
ts
u
s
in
g
FB
SM
with
a
s
et
o
f
tim
e
-
v
ar
y
i
n
g
b
an
d
-
p
ass
r
ec
u
r
s
iv
e
f
ilter
s
o
f
o
r
d
e
r
2
5
w
ith
co
m
p
lem
e
n
tar
y
m
ag
n
itu
d
e
r
esp
o
n
s
es.
T
h
is
all
o
ws
th
e
s
p
ee
ch
s
ig
n
al
to
b
e
p
r
esen
ted
d
ich
o
tically
.
T
h
e
r
ec
u
r
s
iv
e
f
ilter
in
th
e
p
r
o
p
o
s
ed
s
ch
em
e
u
s
es
less
o
r
d
er
h
en
ce
d
ec
r
ea
s
es
p
r
o
ce
s
s
in
g
d
elay
an
d
h
a
r
d
war
e
co
m
p
le
x
ities
co
m
p
ar
ed
to
h
ea
r
in
g
ai
d
s
with
FIR
f
ilter
.
T
h
e
Yu
le
-
W
alk
er
a
p
p
r
o
ac
h
[
2
0
]
is
u
tili
ze
d
to
d
ev
elo
p
th
e
f
ilter
s
,
an
d
th
e
MA
T
L
AB
p
r
o
g
r
am
m
in
g
lan
g
u
ag
e
is
u
s
ed
to
im
p
lem
en
t
th
em
.
E
ac
h
b
an
d
-
p
ass
f
ilter
is
b
ased
o
n
1
/3
o
ctav
e
b
an
d
s
(
to
tal
2
2
b
an
d
s
)
[
2
1
]
.
T
h
e
Yu
le
-
W
alk
er
m
eth
o
d
o
f
d
esig
n
in
g
I
I
R
f
ilter
s
is
u
s
ed
b
ec
au
s
e
ar
b
itra
r
y
m
ag
n
itu
d
e
r
esp
o
n
s
e
ca
n
b
e
d
e
s
ig
n
ed
with
th
is
m
eth
o
d
.
T
h
e
s
y
s
tem
f
u
n
ctio
n
o
f
I
I
R
f
ilter
is
g
iv
en
as in
(
1
)
.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N:
2
2
5
2
-
8
9
3
8
I
n
t J Ar
tif
I
n
tell
,
Vo
l.
1
3
,
No
.
4
,
Dec
em
b
er
2
0
2
4
:
4
9
9
8
-
5
0
0
4
5000
H
(
Z
)
=
∑
−
1
0
−
1
+
∑
−
1
1
−
(
1
)
Her
e
N
is
o
r
d
er
,
a
an
d
b
ar
e
r
esp
ec
tiv
ely
p
o
les
an
d
ze
r
o
s
o
f
th
e
f
ilter
.
|H
m
(
Z
)
|
is
th
e
m
ag
n
itu
d
e
r
esp
o
n
s
e
o
f
tim
e
-
v
a
r
y
in
g
I
I
R
f
ilter
wh
ich
is
d
y
n
am
ic
.
Fig
u
r
e
2
is
s
h
o
w
s
a
s
et
o
f
tim
e
-
v
a
r
y
in
g
b
an
d
-
p
ass
I
I
R
f
ilter
s
'
co
m
p
lem
en
tar
y
m
ag
n
itu
d
e
r
esp
o
n
s
es
f
o
r
Fig
u
r
e
2
(
a)
lef
t
ea
r
an
d
f
o
r
Fig
u
r
e
2
(
b
)
r
ig
h
t
ea
r
ar
e
d
ep
icte
d
at
v
ar
io
u
s
p
o
in
ts
o
f
tim
e.
T
o
r
ed
u
ce
p
h
ase
d
is
to
r
tio
n
in
tr
o
d
u
ce
d
b
y
t
h
e
f
ilter
all
p
ass
f
ilter
is
u
s
ed
[
2
2
]
,
[
2
3
]
.
Fig
u
r
e
1
.
Sch
em
atic
r
ep
r
esen
t
atio
n
u
s
in
g
FB
SM
(
a)
(
b
)
Fig
u
r
e
2
.
Set o
f
tim
e
-
v
ar
y
in
g
b
an
d
-
p
ass
I
I
R
f
ilter
s
'
co
m
p
le
m
en
tar
y
m
a
g
n
itu
d
e
r
esp
o
n
s
es
f
o
r
:
(
a)
lef
t e
ar
an
d
(
b
)
r
ig
h
t e
ar
at
v
ar
io
u
s
p
o
in
ts
o
f
tim
e
3.
RE
SU
L
T
S AN
D
D
I
SCU
SS
I
O
N
I
n
th
is
ap
p
r
o
ac
h
,
a
s
p
ee
ch
s
ig
n
al
u
n
d
er
g
o
es
s
p
ec
tr
al
s
ep
ar
atio
n
th
r
o
u
g
h
th
e
ap
p
licatio
n
o
f
a
s
et
o
f
tim
e
-
v
ar
y
in
g
b
a
n
d
-
p
ass
f
ilter
s
with
co
n
tin
u
o
u
s
s
wee
p
s
in
m
a
g
n
itu
d
e
r
esp
o
n
s
es.
T
h
e
tim
e
s
h
if
ts
b
etwe
en
th
ese
f
ilter
s
ar
e
k
ep
t
s
m
aller
th
an
th
e
ju
s
t
n
o
ticea
b
le
d
if
f
er
e
n
ce
(
J
ND)
,
en
s
u
r
in
g
s
ea
m
less
tr
an
s
itio
n
s
b
etwe
en
s
p
ec
tr
al
co
m
p
o
n
e
n
ts
.
T
h
e
r
esu
ltin
g
s
ig
n
als,
ch
ar
ac
ter
ized
b
y
co
m
p
lem
en
tar
y
s
p
ec
tr
al
co
m
p
o
n
e
n
ts
,
ar
e
th
en
ev
alu
ated
f
o
r
q
u
ality
o
f
v
o
ice
s
ig
n
al
u
s
in
g
PESQ
an
d
MO
S
m
etr
ics,
wh
ile
in
tellig
ib
ilit
y
i
s
ass
es
s
ed
th
r
o
u
g
h
th
e
u
s
e
o
f
th
e
MRT
m
ea
s
u
r
e.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
n
t J Ar
tif
I
n
tell
I
SS
N:
2
2
5
2
-
8
9
3
8
S
p
ec
tr
a
l sp
litt
in
g
o
f sp
ee
ch
s
ig
n
a
l u
s
in
g
time
va
r
yin
g
r
ec
u
r
s
i
ve
filt
ers
fo
r
b
in
a
u
r
a
l …
(
A
p
a
r
n
a
C
h
ila
ka
w
a
d
)
5001
3
.
1
.
Sp
ee
ch
qu
a
lity
3
.
1
.
1
.
P
er
ce
ptua
l e
v
a
lua
t
i
o
n o
f
s
peec
h qua
lity
T
h
e
PESQ
m
o
d
el,
wh
ich
em
p
lo
y
s
th
e
PESQ
alg
o
r
ith
m
to
e
v
alu
ate
s
p
ee
ch
q
u
ality
,
is
r
ep
r
esen
ted
in
Fig
u
r
e
3
.
An
I
n
te
r
n
atio
n
al
T
el
ec
o
m
m
u
n
icatio
n
Un
io
n
s
tan
d
a
r
d
(
I
T
U
-
T
P.8
6
2
)
f
o
r
ev
alu
ati
n
g
s
p
ee
ch
q
u
ality
is
ca
lled
PESQ.
PE
SQ
v
alu
es
ar
e
g
en
er
ated
b
y
an
alg
o
r
ith
m
th
at
r
esem
b
les
a
h
u
m
an
s
en
s
o
r
y
s
y
s
tem
.
T
o
d
eter
m
in
e
th
e
PESQ
v
alu
es
o
f
th
e
p
r
o
ce
s
s
ed
s
ig
n
al,
th
e
PESQ
alg
o
r
ith
m
u
s
es
a
r
ef
er
en
ce
s
ig
n
al
i.e
.
an
u
n
p
r
o
ce
s
s
ed
s
ig
n
al,
an
d
a
p
r
o
ce
s
s
ed
s
ig
n
al
i.e
.
te
s
t
s
ig
n
al
as
in
p
u
ts
.
Stu
d
y
m
ater
ials
th
e
VC
s
y
llab
le
s
/aa
-
b
/
an
d
wo
r
d
/g
r
e
y
/
ar
e
c
o
n
s
id
er
ed
to
ass
ess
th
e
q
u
ality
o
f
th
e
s
p
ee
ch
.
T
h
e
PESQ
v
a
lu
es
(
u
n
p
r
o
ce
s
s
ed
,
p
r
o
ce
s
s
ed
)
o
f
th
e
/aa
-
b
/
an
d
/
g
r
ey
/
s
ig
n
als
at
v
ar
io
u
s
SNR
s
ar
e
g
iv
en
in
T
a
b
les
1
a
n
d
2
,
r
esp
ec
tiv
ely
.
T
h
ese
f
in
d
in
g
s
ar
e
ac
q
u
ir
ed
u
s
in
g
t
h
e
PESQ
alg
o
r
ith
m
,
wh
ich
c
alcu
lates
PESQ
v
alu
es
u
s
in
g
b
o
th
u
n
p
r
o
ce
s
s
ed
(
r
ef
er
en
ce
s
ig
n
al)
an
d
p
r
o
ce
s
s
ed
(
test
s
ig
n
al)
s
ig
n
als.
On
co
m
p
ar
in
g
th
e
PESQ
v
alu
e
s
o
f
p
r
o
ce
s
s
ed
an
d
u
n
p
r
o
ce
s
s
ed
s
ig
n
als,
p
r
o
ce
s
s
e
d
s
ig
n
als
s
h
o
w
im
p
r
o
v
em
en
t
a
t
lo
wer
SNR
s
.
T
h
er
e
is
an
im
p
r
o
v
em
en
t
in
PESQ
v
alu
es
f
o
r
/aa
-
b
/
f
o
r
SNR
v
alu
es
b
etwe
en
6
d
B
an
d
-
6
d
B
in
s
tep
s
o
f
3
d
B
ar
e
0
.
0
9
2
7
,
0
.
1
2
6
1
,
0
.
1
8
3
6
,
0
.
4
6
3
6
,
0
.
3
9
7
9
r
esp
ec
tiv
ely
,
an
d
f
o
r
w
o
r
d
/g
r
ey
/
ar
e
0
.
0
0
2
5
,
0
.
2
6
,
0
.
4
6
9
9
,
0
.
8
3
4
8
,
1
.
1
2
5
9
r
esp
ec
tiv
ely
.
At
lo
w
v
alu
es
o
f
SNR
,
im
p
r
o
v
em
e
n
ts
in
PESQ
v
alu
es a
r
e
o
b
s
er
v
e
d
to
b
e
m
o
r
e.
Fig
u
r
e
3
.
PESQ
m
o
d
el
[
2
4
]
T
ab
le
1
.
Sco
r
es (
PESQ)
f
o
r
V
C
s
y
llab
le
/aa
-
b/
S
N
R
v
a
l
u
e
s
i
n
d
B
/
a
a
-
b
/
si
g
n
a
l
U
n
p
r
o
c
e
sse
d
P
r
o
c
e
ss
e
d
I
mp
r
o
v
e
m
e
n
t
.
∞
4
.
5
2
.
3
5
9
8
-
2
.
1
4
0
2
6
2
.
0
5
7
3
2
.
1
5
0
.
0
9
2
7
3
2
.
0
6
9
4
2
.
1
9
5
5
0
.
1
2
6
1
0
2
.
0
0
5
5
2
.
1
8
9
1
0
.
1
8
3
6
-
3
1
.
9
8
3
9
2
.
4
4
7
5
0
.
4
6
3
6
-
6
1
.
9
1
2
9
2
.
3
1
0
8
0
.
3
9
7
9
T
ab
le
2
.
Sco
r
es (
PESQ)
f
o
r
w
o
r
d
/g
r
ey
/
S
N
R
v
a
l
u
e
s
i
n
d
B
/
g
r
e
y
/
si
g
n
a
l
U
n
p
r
o
c
e
sse
d
P
r
o
c
e
ss
e
d
I
mp
r
o
v
e
m
e
n
t
.
∞
4
.
5
1
.
6
6
-
2
.
8
4
6
1
.
8
2
6
4
1
.
8
2
8
9
0
.
0
0
2
5
3
1
.
5
9
7
1
.
8
5
7
0
.
2
6
0
1
.
4
4
4
9
1
.
9
1
4
8
0
.
4
6
9
9
-
3
1
.
3
1
0
1
2
.
1
4
4
9
0
.
8
3
4
8
-
6
1
.
2
2
4
3
2
.
3
5
0
2
1
.
1
2
5
9
3
.
1
.
2
.
M
ea
n o
pin
io
n sco
re
An
ass
ess
m
en
t
o
f
s
p
ee
ch
q
u
ality
is
also
m
ad
e
u
s
in
g
t
h
e
MO
S
m
ea
s
u
r
e
o
n
th
r
ee
m
ale
an
d
th
r
ee
f
em
ale
s
u
b
jects
ag
ed
b
etwe
en
2
1
to
4
5
y
ea
r
s
with
n
o
r
m
a
l
h
ea
r
in
g
.
W
ith
in
itial
tr
ain
in
g
in
th
e
ass
ess
m
en
t
m
eth
o
d
,
p
ar
ticip
an
ts
wer
e
ask
ed
to
p
r
o
v
id
e
t
h
eir
o
p
in
io
n
(
o
n
a
s
ca
le
0
-
5
,
b
ad
-
1
,
p
o
o
r
-
2
,
f
air
-
3
,
g
o
o
d
-
4
,
an
d
ex
ce
llen
t
-
5
[
2
5
]
)
a
b
o
u
t
t
h
e
q
u
al
it
y
o
f
u
n
p
r
o
c
ess
e
d
a
n
d
p
r
o
ce
s
s
e
d
m
ate
r
i
als
at
v
a
r
i
o
u
s
S
NR
le
v
els.
T
h
e
f
i
r
s
t
co
l
u
m
n
o
f
b
o
t
h
T
a
b
les
3
a
n
d
4
lis
ts
d
i
f
f
e
r
e
n
t
s
u
b
jec
ts
al
o
n
g
wi
th
t
h
eir
ag
e
a
n
d
s
e
x
.
T
h
e
f
o
ll
o
w
in
g
c
o
l
u
m
n
s
p
r
ese
n
t
t
h
ei
r
o
p
i
n
i
o
n
s
b
ase
d
o
n
t
h
e
p
e
r
ce
iv
e
d
q
u
ali
ty
o
f
b
o
t
h
u
n
p
r
o
c
ess
e
d
a
n
d
p
r
o
ce
s
s
ed
m
at
er
ials
at
v
ar
io
u
s
SNR
le
v
e
ls
(
i
n
f
in
it
y
,
6
,
3
,
0
,
-
3
,
a
n
d
-
6
d
B
)
.
B
y
d
et
er
m
i
n
i
n
g
t
h
e
a
v
e
r
ag
e
o
f
o
p
in
io
n
s
c
o
r
es,
MO
S
v
a
lu
es
ar
e
o
b
tai
n
ed
.
I
m
p
r
o
v
em
e
n
t c
a
n
b
e
s
ee
n
in
MO
S
v
al
u
es o
f
p
r
o
ce
s
s
ed
test
m
ate
r
i
als at
lo
we
r
SN
R
o
n
c
o
m
p
a
r
i
n
g
t
h
e
MO
S
o
f
u
n
p
r
o
ce
s
s
ed
s
i
g
n
a
l
wit
h
p
r
o
ce
s
s
ed
.
A
n
e
n
h
a
n
c
em
en
t
in
M
OS
v
al
u
es
f
o
r
t
h
e
V
C
s
i
g
n
al
/
aa
-
b
/
a
f
te
r
p
r
o
ce
s
s
i
n
g
is
0
.
5
8
,
0
.
9
2
,
1
.
0
0
,
1
.
0
8
,
a
n
d
1
.
1
7
f
o
r
S
NR
6
t
o
-
6
i
n
s
t
ep
s
o
f
3
d
B
r
es
p
e
cti
v
e
ly
.
Si
m
il
ar
t
o
t
h
is
,
MO
S
is
u
s
e
d
to
e
x
a
m
i
n
e
th
e
v
o
ic
e
q
u
al
it
y
o
f
th
e
w
o
r
d
/
g
r
ey
/
o
n
t
h
e
s
a
m
e
p
a
r
t
ici
p
a
n
ts
,
an
d
th
e
r
esu
lts
s
h
o
we
d
an
im
p
r
o
v
em
en
t
f
o
r
SNR
6
t
o
-
6
i
n
s
te
p
s
o
f
3
d
B
is
0
.
6
7
,
0
.
9
2
,
1
.
0
8
,
1
.
1
7
,
a
n
d
1
.
2
5
r
es
p
ec
ti
v
el
y
(
p
r
o
c
ess
e
d
s
p
e
ec
h
)
.
Fo
r
b
o
t
h
tes
t s
ig
n
a
ls
,
p
r
o
ce
s
s
e
d
s
i
g
n
als
h
a
v
e
h
ig
h
er
MO
S s
co
r
es
c
o
m
p
a
r
e
d
to
u
n
p
r
o
c
ess
e
d
s
ig
n
a
ls
.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N:
2
2
5
2
-
8
9
3
8
I
n
t J Ar
tif
I
n
tell
,
Vo
l.
1
3
,
No
.
4
,
Dec
em
b
er
2
0
2
4
:
4
9
9
8
-
5
0
0
4
5002
T
ab
le
3
.
Sco
r
es (
MO
S)
f
o
r
u
n
p
r
o
ce
s
s
ed
an
d
p
r
o
ce
s
s
ed
/aa
-
b
/ sig
n
al
S
u
b
j
e
c
t
s w
i
t
h
s
e
x
,
a
g
e
S
N
R
i
n
d
B
∞
6
3
0
-
3
-
6
*
U
n
p
r
**
Pro
U
n
p
r
Pro
U
n
p
r
Pro
U
n
p
r
Pro
U
n
p
r
Pro
U
n
p
r
Pro
S
1
,
[
M
,
2
0
]
4
.
5
3
2
.
5
3
2
.
5
3
2
.
5
3
2
.
5
3
.
5
2
3
.
5
S
2
,
[
F
,
2
5
]
4
.
5
2
.
5
2
.
5
3
.
5
2
3
2
2
.
5
2
3
1
2
S
3
,
[
F
,
2
1
]
4
3
3
3
1
.
5
2
.
5
2
.
5
3
.
5
2
.
5
3
.
5
2
3
S
4
,
[
F
,
3
3
]
4
3
2
.
5
3
2
3
.
5
2
3
.
5
2
3
1
2
S
5
,
[
M
,
4
5
]
4
3
.
5
2
3
.
5
1
.
5
2
.
5
1
2
.
5
1
2
.
5
1
2
.
5
S
6
,
[
M
,
2
7
]
4
2
.
5
3
3
2
2
.
5
2
3
2
3
2
3
M
O
S
4
.
1
7
2
.
9
2
2
.
5
8
3
.
1
7
1
.
9
2
2
.
8
3
2
.
0
0
3
.
0
0
2
.
0
0
3
.
0
8
1
.
5
0
2
.
6
7
I
mp
r
o
v
e
m
e
n
t
-
1
.
2
5
0
.
5
8
0
.
9
2
1
.
0
0
1
.
0
8
1
.
1
7
*
U
n
p
r
:
U
n
p
r
o
c
e
ss
e
d
sp
e
e
c
h
,
**
P
r
o
:
P
r
o
c
e
sse
d
s
p
e
e
c
h
T
ab
le
4
.
Sco
r
es (
MO
S)
f
o
r
u
n
p
r
o
ce
s
s
ed
an
d
p
r
o
ce
s
s
ed
/g
r
e
y
/ sig
n
al
S
u
b
j
e
c
t
s w
i
t
h
s
e
x
,
a
g
e
S
N
R
i
n
d
B
∞
6
3
0
-
3
-
6
U
n
p
r
Pro
U
n
p
r
Pro
U
n
p
r
Pro
U
n
p
r
Pro
U
n
p
r
Pro
U
n
p
r
Pro
S
1
,
[
M
,
2
0
]
4
.
5
3
2
.
5
3
2
2
.
5
2
.
5
3
2
.
5
3
2
3
S
2
,
[
F
,
2
5
]
4
2
.
5
2
3
.
5
2
3
2
2
.
5
2
3
.
5
1
2
.
5
S
3
,
[
F
,
2
1
]
4
3
3
3
.
5
1
.
5
2
.
5
2
.
5
3
.
5
2
.
5
3
.
5
1
2
.
5
S
4
,
[
F
,
3
3
]
4
3
2
.
5
3
2
3
2
3
.
5
2
3
2
2
.
5
S
5
,
[
M
,
4
5
]
4
.
5
3
2
3
.
5
1
2
.
5
1
2
.
5
1
3
1
3
S
6
,
[
M
,
2
7
]
4
2
.
5
3
.
5
3
1
.
5
2
2
3
.
5
2
3
2
3
M
O
S
4
.
1
7
2
.
8
3
2
.
5
8
3
.
2
5
1
.
6
7
2
.
5
8
2
.
0
0
3
.
0
8
2
.
0
0
3
.
1
7
1
.
5
0
2
.
7
5
I
mp
r
o
v
e
m
e
n
t
-
1
.
3
3
0
.
6
7
0
.
9
2
1
.
0
8
1
.
1
7
1
.
2
5
3
.
2
.
Sp
ee
ch
inte
llig
ibi
lity
T
h
e
MRT
is
u
s
ed
t
o
ass
ess
s
p
ee
ch
in
tellig
ib
ilit
y
.
3
0
0
co
n
s
o
n
an
t
-
v
o
wel
-
co
n
s
o
n
a
n
t
(
C
VC
)
wo
r
d
s
m
ad
e
u
p
th
e
s
tu
d
y
m
ater
ials
f
o
r
MRT.
E
ac
h
wo
r
d
is
p
r
ec
e
d
ed
b
y
"wo
u
l
d
y
o
u
wr
ite".
T
h
e
wo
r
d
s
wer
e
u
tter
ed
b
y
a
g
u
y
in
a
n
au
d
io
m
et
r
y
r
o
o
m
an
d
r
ec
o
r
d
ed
at
1
0
k
ilo
s
am
p
les
p
er
s
ec
.
with
1
6
-
b
it
q
u
an
tizatio
n
u
s
in
g
a
B
&
K
m
icr
o
p
h
o
n
e
m
o
d
el
No
.
2
,
2
1
0
.
Six
test
s
ets
o
f
5
0
wo
r
d
s
ea
ch
wer
e
m
ad
e
f
r
o
m
all
3
0
0
C
VC
wo
r
d
s
.
All
test
wo
r
d
s
wer
e
ch
o
s
en
u
s
in
g
a
two
-
lev
el
r
an
d
o
m
izatio
n
p
r
o
ce
s
s
to
o
v
er
c
o
m
e
b
iasi
n
g
.
F
o
r
lis
ten
in
g
test
s
to
d
eter
m
in
e
in
tellig
ib
ilit
y
,
s
u
b
je
cts
with
m
o
d
er
ate
b
ilater
al
S
NHL
(
s
ix
in
n
u
m
b
e
r
,
3
m
en
an
d
3
wo
m
e
n
,
ag
es
3
2
to
5
1
)
p
ar
ticip
ated
.
T
h
e
s
u
b
jects
wer
e
f
ir
s
t
g
iv
en
d
ir
ec
ti
o
n
s
o
n
th
e
test
in
g
p
r
o
ce
s
s
an
d
s
tim
u
li.
On
ce
th
ey
wer
e
u
s
ed
to
th
e
test
in
g
p
r
o
ce
s
s
,
th
e
C
VC
wo
r
d
s
wer
e
p
r
ese
n
ted
to
th
em
.
T
h
e
wh
o
le
test
in
g
p
r
o
ce
s
s
was d
o
n
e
u
s
in
g
an
au
t
o
m
ated
test
ad
m
in
is
tr
atio
n
s
etu
p
.
I
n
th
e
test
in
g
p
r
o
ce
s
s
,
1
,
8
0
0
w
o
r
d
s
(
3
0
0
wo
r
d
s
x
6
le
v
els o
f
SNR
)
wer
e
p
r
esen
ted
to
ea
ch
p
ar
ticip
an
t
to
r
ep
ly
to
.
I
n
th
e
MRT
test
win
d
o
w,
3
0
0
MRT
wo
r
d
s
we
r
e
lo
ad
ed
a
n
d
cr
ea
te
d
a
r
an
d
o
m
f
ile
to
s
to
r
e
th
e
r
ec
o
g
n
itio
n
s
co
r
e
o
f
p
ar
ticip
a
n
ts
.
E
ac
h
s
u
b
ject
was
in
s
tr
u
cted
to
s
it
in
f
r
o
n
t
o
f
th
e
c
o
m
p
u
ter
s
cr
ee
n
,
an
d
to
click
th
e
test
win
d
o
w'
s
"p
lay
"
b
u
tto
n
.
T
h
e
s
ix
o
p
tio
n
s
p
r
esen
ted
in
th
e
test
win
d
o
w,
th
e
s
u
b
ject
ch
o
s
e
th
e
b
est
m
atch
af
ter
h
ea
r
in
g
th
e
p
r
esen
tatio
n
.
R
esp
o
n
s
e
tim
e
(
s
ec
)
a
n
d
s
p
ee
ch
r
ec
o
g
n
itio
n
s
co
r
e
(
%
)
wer
e
r
ec
o
r
d
ed
in
a
r
an
d
o
m
f
ile.
T
h
is
p
r
o
ce
d
u
r
e
was
d
o
n
e
f
o
r
all
MRT
wo
r
d
s
with
d
if
f
er
e
n
t
SNR
lev
els.
Dep
en
d
in
g
o
n
th
e
p
ar
ticip
an
ts
’
av
ailab
ilit
y
an
d
willin
g
n
ess
,
th
e
test
was
r
u
n
f
o
r
ar
o
u
n
d
a
m
o
n
th
.
T
h
e
au
d
i
o
m
etr
ic
th
r
esh
o
ld
f
o
r
th
e
p
ar
ticip
an
ts
with
b
ilater
al
m
o
d
er
ate
SNHL
is
g
iv
en
in
T
ab
le
5
.
T
ab
le
5
.
T
h
r
esh
o
ld
le
v
el
f
o
r
th
e
h
u
m
an
s
with
SNHL
(
m
o
d
er
a
te
b
ilater
al)
in
au
d
i
o
g
r
am
S
u
b
j
e
c
t
Ea
r
H
e
a
r
i
n
g
t
h
r
e
s
h
o
l
d
l
e
v
e
l
(
d
B
H
L)
F
r
e
q
u
e
n
c
y
2
5
0
H
z
5
0
0
H
z
1
0
0
0
H
z
2
0
0
0
H
z
4
0
0
0
H
z
8
0
0
0
H
z
S
1
[
M
,
4
2
]
*
R
60
45
55
40
60
65
**
L
40
55
55
50
60
60
S
2
[
M
,
5
1
]
R
50
55
50
45
55
60
L
55
50
45
55
50
55
S
3
[
M
,
4
3
]
R
20
30
60
40
50
50
L
20
30
60
50
40
50
S
4
[
F
,
3
5
]
R
65
60
55
50
40
45
L
60
55
55
50
45
50
S
5
[
F
,
3
2
]
R
25
30
40
50
60
90
L
20
30
40
50
60
85
S
6
[
F
,
4
0
]
R
35
45
55
60
65
70
L
25
40
45
50
60
60
*
R
:
R
i
g
h
t
e
a
r
,
**
L:
Le
f
t
e
a
r
Evaluation Warning : The document was created with Spire.PDF for Python.
I
n
t J Ar
tif
I
n
tell
I
SS
N:
2
2
5
2
-
8
9
3
8
S
p
ec
tr
a
l sp
litt
in
g
o
f sp
ee
ch
s
ig
n
a
l u
s
in
g
time
va
r
yin
g
r
ec
u
r
s
i
ve
filt
ers
fo
r
b
in
a
u
r
a
l …
(
A
p
a
r
n
a
C
h
ila
ka
w
a
d
)
5003
T
h
e
r
esu
lts
o
f
lis
ten
in
g
test
s
c
o
n
d
u
cte
d
o
n
p
ar
ticip
a
n
ts
u
s
in
g
MRT
m
ater
ials
ar
e
s
h
o
wn
in
T
ab
le
6
,
with
th
e
av
e
r
ag
e
au
d
io
r
ec
o
g
n
itio
n
s
co
r
e
(
%)
f
o
r
b
o
t
h
u
n
p
r
o
ce
s
s
ed
an
d
p
r
o
ce
s
s
ed
s
p
ee
ch
at
v
ar
io
u
s
SNR
lev
els.
T
h
e
av
er
ag
e
r
ec
o
g
n
itio
n
s
co
r
es
o
f
p
r
o
ce
s
s
ed
s
p
ee
c
h
h
av
e
im
p
r
o
v
ed
b
y
-
1
.
7
8
%,
1
5
.
0
6
%,
1
8
.
2
2
%,
2
0
.
6
2
%,
2
4
.
1
7
%,
a
n
d
2
6
.
6
1
%
r
esp
ec
tiv
ely
,
wh
e
n
c
o
m
p
ar
e
d
to
th
e
u
n
p
r
o
ce
s
s
ed
r
ec
o
g
n
itio
n
s
co
r
es
with
SNR
lev
els
(
∞
d
B
,
6
d
B
to
+6
d
B
in
in
cr
em
en
ts
o
f
3
d
B
)
.
T
a
b
le
7
s
h
o
ws
th
e
av
er
ag
e
r
esp
o
n
s
e
tim
es
o
f
th
e
s
am
e
s
u
b
jects f
o
r
th
e
s
p
ec
if
ied
SNR
lev
els,
an
d
it c
an
b
e
s
ee
n
th
at
r
esp
o
n
s
e
tim
e
is
d
ec
r
ea
s
ed
f
o
r
p
r
o
ce
s
s
ed
s
p
ee
ch
.
T
ab
le
6
.
Sp
ee
ch
r
ec
o
g
n
itio
n
s
co
r
es (
%)
f
o
r
SNHL
(
b
ilater
al)
s
u
b
jects
S
u
b
j
e
c
t
s w
i
t
h
s
e
x
,
a
g
e
S
N
R
V
a
l
u
e
s
i
n
d
B
∞
6
3
0
-
3
-
6
U
n
p
r
Pro
U
n
p
r
Pro
U
n
p
r
Pro
U
n
p
r
Pro
U
n
p
r
Pro
U
n
p
r
Pro
S
1
[
M
,
4
2
]
9
2
.
0
0
9
1
.
6
7
6
5
.
3
3
8
1
.
3
3
5
7
.
0
0
7
5
.
0
0
5
3
.
6
0
7
7
.
0
0
4
6
.
6
7
6
7
.
0
0
4
2
.
6
7
7
1
.
0
0
S
2
[
M
,
5
1
]
9
2
.
3
3
9
0
.
3
3
6
5
.
3
3
8
0
.
6
7
5
7
.
6
7
7
7
.
0
0
5
2
.
4
0
7
4
.
0
0
4
9
.
3
3
7
8
.
0
0
4
3
.
6
7
7
1
.
3
3
S
3
[
M
,
4
3
]
9
4
.
6
7
9
4
.
3
3
6
6
.
3
3
7
8
.
6
7
5
8
.
6
7
7
9
.
0
0
5
5
.
6
7
7
6
.
3
3
4
9
.
3
3
7
3
.
6
7
4
5
.
3
3
7
2
.
0
0
S
4
[
F
,
3
5
]
9
4
.
0
0
9
2
.
6
7
6
6
.
3
3
8
1
.
6
7
6
1
.
0
0
7
6
.
3
3
5
6
.
0
0
7
6
.
4
0
4
7
.
6
7
7
2
.
3
3
4
7
.
0
0
7
2
.
6
7
S
5
[
F
,
3
2
]
9
2
.
0
0
8
6
.
3
3
6
6
.
3
3
8
0
.
0
0
5
8
.
6
7
7
8
.
0
0
5
5
.
3
3
7
2
.
6
7
4
7
.
6
7
7
3
.
0
0
4
5
.
3
3
7
2
.
3
3
S
6
[
F
,
4
0
]
9
2
.
6
7
9
1
.
6
7
6
5
.
3
3
8
3
.
0
0
6
1
.
0
0
7
8
.
0
0
5
8
.
3
3
7
8
.
6
7
5
0
.
3
3
7
2
.
0
0
4
5
.
3
3
6
9
.
6
7
AVG.
9
2
.
9
4
9
1
.
1
7
6
5
.
8
3
8
0
.
8
9
5
9
.
0
0
7
7
.
2
2
5
5
.
2
2
7
5
.
8
4
4
8
.
5
0
7
2
.
6
7
4
4
.
8
9
7
1
.
5
0
I
mp
r
o
v
e
m
e
n
t
-
1
.
7
8
1
5
.
0
6
1
8
.
2
2
2
0
.
6
2
2
4
.
1
7
2
6
.
6
1
T
ab
le
7
.
R
esp
o
n
s
e
tim
e
in
s
ec
o
n
d
s
f
o
r
b
ilater
al
SNHL
s
u
b
jects
S
u
b
j
e
c
t
R
e
s
p
o
n
se
T
i
me
S
N
R
(
d
B
)
∞
d
B
+
6
d
B
+
3
d
B
0
d
B
-
3
d
B
-
6
d
B
U
n
p
r
Pro
U
n
p
r
Pro
U
n
p
r
Pro
U
n
p
r
Pro
U
n
p
r
Pro
U
n
p
r
Pro
S1
4
.
5
4
3
4
.
5
8
5
4
.
8
6
5
4
.
4
6
5
4
.
4
7
6
4
.
9
8
1
4
.
7
5
4
4
.
8
7
6
4
.
3
1
8
4
.
7
4
3
5
.
7
6
2
4
.
7
6
6
S2
4
.
4
5
6
4
.
6
4
5
4
.
5
4
3
4
.
4
5
4
5
.
4
3
5
4
.
7
6
8
4
.
6
6
5
4
.
5
4
6
4
.
7
6
1
4
.
3
2
4
5
.
7
6
2
4
.
7
2
3
S3
4
.
8
7
9
4
.
7
6
5
4
.
8
5
4
4
.
7
8
6
5
.
9
8
7
4
.
5
4
3
5
.
9
4
3
4
.
9
8
5
5
.
8
7
2
4
.
1
9
8
5
.
1
2
8
4
.
1
8
1
S4
4
.
8
6
5
4
.
9
5
4
4
.
5
4
3
4
.
4
3
5
4
.
3
4
5
4
.
9
8
4
4
.
8
7
4
4
.
7
5
4
4
.
8
2
1
4
.
6
4
3
4
.
9
8
4
4
.
2
4
1
S5
4
.
5
4
2
4
.
5
4
3
4
.
9
4
3
4
.
7
6
5
4
.
8
7
2
4
.
5
6
7
4
.
8
7
6
4
.
8
9
7
4
.
8
4
3
4
.
7
6
5
5
.
8
4
5
4
.
9
4
1
S6
4
.
7
8
6
4
.
3
4
5
4
.
3
4
2
4
.
5
4
3
4
.
7
6
5
4
.
8
9
4
4
.
6
5
4
4
.
6
3
4
4
.
6
5
4
4
.
4
3
5
5
.
7
5
5
4
.
9
5
4
M
e
a
n
4
.
6
7
9
4
.
6
4
0
4
.
6
8
2
4
.
5
7
5
4
.
9
8
0
4
.
7
9
0
4
.
9
6
1
4
.
7
8
2
4
.
8
7
8
4
.
5
1
8
5
.
5
3
9
4
.
6
3
4
R
e
d
u
c
t
i
o
n
0
.
0
4
0
.
1
1
0
.
1
9
0
.
1
8
0
.
3
6
0
.
9
1
4.
CO
NCLU
SI
O
N
T
h
e
t
e
c
h
n
i
q
u
e
p
r
o
p
o
s
e
d
i
n
t
h
i
s
p
a
p
e
r
i
n
v
o
l
v
e
s
d
i
v
i
d
i
n
g
t
h
e
v
o
i
c
e
s
i
g
n
a
l
i
n
t
o
t
w
o
p
a
r
t
i
t
i
o
n
s
w
i
t
h
c
o
m
p
l
e
m
e
n
t
a
r
y
s
p
e
c
t
r
a
l
c
o
m
p
o
n
e
n
t
s
f
o
r
a
b
i
n
a
u
r
a
l
d
i
c
h
o
t
i
c
h
e
a
r
i
n
g
a
i
d
.
T
h
i
s
i
s
c
a
r
r
i
e
d
o
u
t
b
y
u
s
i
n
g
a
s
e
t
o
f
t
i
m
e
-
v
a
r
y
i
n
g
b
a
n
d
-
p
a
s
s
(
I
I
R
)
f
i
l
t
e
r
s
w
i
t
h
o
n
e
-
t
h
i
r
d
o
c
t
a
v
e
b
a
n
d
s
h
a
v
i
n
g
c
o
n
t
i
n
u
o
u
s
s
w
e
e
p
i
n
m
a
g
n
i
t
u
d
e
r
e
s
p
o
n
s
e
s
w
i
t
h
t
i
m
e
s
h
i
f
t
c
h
o
s
e
n
l
e
s
s
t
h
a
n
J
N
D
.
T
h
e
m
e
t
h
o
d
d
e
c
r
e
a
s
e
s
t
h
e
i
m
p
a
c
t
o
f
s
p
e
c
t
r
a
l
m
a
s
k
i
n
g
a
n
d
e
n
h
a
n
c
e
s
t
h
e
a
b
i
l
i
t
y
o
f
p
e
o
p
l
e
w
i
t
h
S
N
H
L
t
o
d
e
t
e
c
t
g
a
p
s
i
n
a
v
o
i
c
e
s
i
g
n
a
l
.
T
h
e
s
c
h
e
m
e
i
s
e
v
a
l
u
a
t
e
d
u
s
i
n
g
m
e
a
s
u
r
e
s
:
P
E
S
Q
,
M
O
S
t
e
s
t
f
o
r
q
u
a
l
i
t
y
,
a
n
d
M
R
T
f
o
r
s
p
e
e
c
h
i
n
t
e
l
l
i
g
i
b
i
l
i
t
y
.
T
h
e
r
e
s
u
l
t
s
o
f
t
h
e
P
E
S
Q
a
n
d
M
O
S
s
t
u
d
i
e
s
i
n
d
i
c
a
t
e
d
a
n
i
m
p
r
o
v
e
m
e
n
t
i
n
t
h
e
P
E
S
Q
v
a
l
u
e
o
f
0
.
3
9
7
9
a
n
d
M
O
S
t
e
s
t
s
c
o
r
e
o
f
1
.
1
7
f
o
r
/
a
a
-
b
/
a
t
S
N
R
-
6
d
B
o
n
c
o
m
p
a
r
i
n
g
u
n
p
r
o
c
e
s
s
e
d
s
p
e
e
c
h
w
i
t
h
p
r
o
c
e
s
s
e
d
s
p
e
e
c
h
.
S
i
m
i
l
a
r
l
y
,
i
m
p
r
o
v
e
m
e
n
t
i
n
P
E
S
Q
v
a
l
u
e
o
f
1
.
1
2
5
9
a
n
d
M
O
S
t
e
s
t
s
c
o
r
e
o
f
1
.
2
5
f
o
r
/
g
r
e
y
/
a
t
-
6
d
B
S
N
R
o
n
c
o
m
p
a
r
i
n
g
u
n
p
r
o
c
e
s
s
e
d
s
p
e
e
c
h
w
i
t
h
p
r
o
c
e
s
s
e
d
s
p
e
e
c
h
.
T
h
e
M
R
T
r
e
s
u
l
t
s
h
o
w
e
d
2
6
.
6
1
%
i
m
p
r
o
v
e
m
e
n
t
i
n
r
e
c
o
g
n
i
t
i
o
n
s
c
o
r
e
a
n
d
a
d
e
c
r
e
a
s
e
d
r
e
s
p
o
n
s
e
t
i
m
e
o
f
0
.
9
1
s
e
c
i
n
t
h
e
c
a
s
e
o
f
b
i
l
a
t
e
r
a
l
S
N
H
L
p
e
r
s
o
n
s
w
h
e
n
p
r
o
c
e
s
s
e
d
s
p
e
e
c
h
i
s
c
o
m
p
a
r
e
d
w
i
t
h
t
h
a
t
o
f
u
n
p
r
o
c
e
s
s
e
d
s
p
e
e
c
h
a
t
t
h
e
s
a
m
e
v
a
l
u
e
o
f
S
N
R
.
T
h
e
q
u
a
l
i
t
y
a
n
d
i
n
t
e
l
l
i
g
i
b
i
l
i
t
y
o
f
p
r
o
c
e
s
s
e
d
s
p
e
e
c
h
(
p
r
o
c
e
s
s
e
d
u
s
i
n
g
t
h
e
p
r
o
p
o
s
e
d
s
c
h
e
m
e
)
h
a
v
e
i
m
p
r
o
v
e
d
m
o
r
e
a
t
l
o
w
e
r
S
N
R
i
m
p
l
i
e
s
s
p
e
e
c
h
p
e
r
c
e
p
t
i
o
n
i
n
c
r
e
a
s
e
s
i
n
S
N
H
L
h
u
m
a
n
s
.
H
e
n
c
e
t
h
e
p
r
o
p
o
s
e
d
m
e
t
h
o
d
i
s
m
o
r
e
e
f
f
e
c
t
i
v
e
u
n
d
e
r
a
d
v
e
r
s
e
l
i
s
t
e
n
i
n
g
c
o
n
d
i
t
i
o
n
s
.
I
n
f
u
t
u
r
e
m
o
r
e
n
u
m
b
e
r
o
f
p
a
r
t
i
c
i
p
a
n
t
s
a
n
d
m
u
l
t
i
b
a
n
d
c
o
m
p
r
e
s
s
i
o
n
r
e
q
u
i
r
e
t
o
b
e
c
o
n
s
i
d
e
r
e
d
.
RE
F
E
R
E
NC
E
S
[
1
]
J.
E.
C
.
L
i
e
u
,
M
.
K
e
n
n
a
,
S
.
A
n
n
e
,
a
n
d
L.
D
a
v
i
d
s
o
n
,
“
H
e
a
r
i
n
g
l
o
ss
i
n
c
h
i
l
d
r
e
n
:
a
r
e
v
i
e
w
,
”
J
AM
A
-
J
o
u
r
n
a
l
o
f
t
h
e
Am
e
r
i
c
a
n
Me
d
i
c
a
l
Ass
o
c
i
a
t
i
o
n
,
v
o
l
.
3
2
4
,
n
o
.
2
1
,
p
p
.
2
1
9
5
–
2
2
0
5
,
2
0
2
0
,
d
o
i
:
1
0
.
1
0
0
1
/
j
a
ma.
2
0
2
0
.
1
7
6
4
7
.
[
2
]
R
.
S
.
P
u
j
a
r
a
n
d
P
.
N
.
K
u
l
k
a
r
n
i
,
“
C
a
sc
a
d
e
d
s
t
r
u
c
t
u
r
e
o
f
n
o
i
se
r
e
d
u
c
t
i
o
n
a
n
d
mu
l
t
i
b
a
n
d
f
r
e
q
u
e
n
c
y
c
o
mp
r
e
ssi
o
n
o
f
sp
e
e
c
h
si
g
n
a
l
t
o
i
mp
r
o
v
e
s
p
e
e
c
h
p
e
r
c
e
p
t
i
o
n
f
o
r
mo
n
a
u
r
a
l
h
e
a
r
i
n
g
a
i
d
s
,
”
2
0
1
9
I
EEE
1
6
t
h
I
n
d
i
a
C
o
u
n
c
i
l
I
n
t
e
r
n
a
t
i
o
n
a
l
C
o
n
f
e
re
n
c
e
,
I
N
D
I
C
O
N
2
0
1
9
-
S
y
m
p
o
s
i
u
m
Pr
o
c
e
e
d
i
n
g
s
,
2
0
1
9
,
d
o
i
:
1
0
.
1
1
0
9
/
I
N
D
I
C
O
N
4
7
2
3
4
.
2
0
1
9
.
9
0
3
0
3
6
6
.
[
3
]
R
.
S
.
P
u
j
a
r
a
n
d
P
.
N
.
K
u
l
k
a
r
n
i
,
“
W
i
e
n
e
r
f
i
l
t
e
r
-
b
a
se
d
n
o
i
s
e
r
e
d
u
c
t
i
o
n
a
l
g
o
r
i
t
h
m
w
i
t
h
p
e
r
c
e
p
t
u
a
l
p
o
st
f
i
l
t
e
r
i
n
g
f
o
r
h
e
a
r
i
n
g
a
i
d
s,
”
I
n
t
e
r
n
a
t
i
o
n
a
l
J
o
u
r
n
a
l
o
f
I
m
a
g
e
,
G
ra
p
h
i
c
s
a
n
d
S
i
g
n
a
l
P
ro
c
e
ss
i
n
g
,
v
o
l
.
1
1
,
n
o
.
7
,
p
p
.
6
9
–
8
1
,
2
0
1
9
,
d
o
i
:
1
0
.
5
8
1
5
/
i
j
i
g
sp
.
2
0
1
9
.
0
7
.
0
6
.
[
4
]
A
.
N
.
C
h
e
e
r
a
n
,
P
.
C
.
P
a
n
d
e
y
,
a
n
d
D
.
S
.
Jan
g
a
mas
h
e
t
t
i
s,
“
D
e
si
g
n
o
f
c
o
mb
f
i
l
t
e
r
s
b
a
se
d
o
n
a
u
d
i
t
o
r
y
f
i
l
t
e
r
b
a
n
d
w
i
d
t
h
s
f
o
r
b
i
n
a
u
r
a
l
d
i
c
h
o
t
i
c
p
r
e
se
n
t
a
t
i
o
n
f
o
r
p
e
r
s
o
n
s
w
i
t
h
se
n
s
o
r
i
n
e
u
r
a
l
h
e
a
r
i
n
g
i
mp
a
i
r
me
n
t
,
”
I
n
t
e
rn
a
t
i
o
n
a
l
C
o
n
f
e
re
n
c
e
o
n
D
i
g
i
t
a
l
S
i
g
n
a
l
Pro
c
e
ss
i
n
g
,
D
S
P
,
v
o
l
.
2
,
p
p
.
9
7
1
–
9
7
4
,
2
0
0
2
,
d
o
i
:
1
0
.
1
1
0
9
/
I
C
D
S
P
.
2
0
0
2
.
1
0
2
8
2
5
2
.
[
5
]
P
.
N
.
K
u
l
k
a
r
n
i
a
n
d
P
.
C
.
P
a
n
d
e
y
,
“
O
p
t
i
m
i
z
i
n
g
t
h
e
c
o
m
b
f
i
l
t
e
r
s
f
o
r
s
p
e
c
t
r
a
l
s
p
l
i
t
t
i
n
g
o
f
s
p
e
e
c
h
t
o
r
e
d
u
c
e
t
h
e
e
f
f
e
c
t
o
f
s
p
e
c
t
r
a
l
m
a
s
k
i
n
g
,
”
I
n
t
e
r
n
a
t
i
o
n
a
l
C
o
n
f
e
r
e
n
c
e
o
n
S
i
g
n
a
l
P
r
o
c
e
s
s
i
n
g
C
o
m
m
u
n
i
c
a
t
i
o
n
s
a
n
d
N
e
t
w
o
r
k
i
n
g
,
p
p
.
6
9
–
7
3
,
2
0
0
8
,
d
o
i
:
1
0
.
1
1
0
9
/
I
C
S
C
N
.
2
0
0
8
.
4
4
4
7
1
6
3
.
[
6
]
S
.
G
.
K
a
m
b
a
l
i
ma
t
h
,
P
.
N
.
K
u
l
k
a
r
n
i
,
P
.
C
.
P
a
n
d
e
y
,
S
.
S
.
M
.
-
S
h
e
t
t
i
,
a
n
d
S
.
G
.
H
i
r
e
m
a
t
h
,
“
F
P
G
A
b
a
s
e
d
i
m
p
l
e
m
e
n
t
a
t
i
o
n
o
f
c
o
m
b
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N:
2
2
5
2
-
8
9
3
8
I
n
t J Ar
tif
I
n
tell
,
Vo
l.
1
3
,
No
.
4
,
Dec
em
b
er
2
0
2
4
:
4
9
9
8
-
5
0
0
4
5004
f
i
l
t
e
r
s
f
o
r
u
se
i
n
b
i
n
a
u
r
a
l
h
e
a
r
i
n
g
a
i
d
s
f
o
r
r
e
d
u
c
i
n
g
i
n
t
r
a
s
p
e
e
c
h
s
p
e
c
t
r
a
l
m
a
sk
i
n
g
,
”
2
0
1
3
A
n
n
u
a
l
I
EEE
I
n
d
i
a
C
o
n
f
e
re
n
c
e
,
I
N
D
I
C
O
N
2
0
1
3
,
2
0
1
3
,
d
o
i
:
1
0
.
1
1
0
9
/
I
N
D
C
O
N
.
2
0
1
3
.
6
7
2
6
0
1
2
.
[
7
]
P
.
P
u
j
a
r
,
A
.
K
u
m
a
r
,
a
n
d
V
.
K
u
mar
,
“
Ef
f
i
c
i
e
n
t
p
l
a
n
t
l
e
a
f
d
e
t
e
c
t
i
o
n
t
h
r
o
u
g
h
ma
c
h
i
n
e
l
e
a
r
n
i
n
g
a
p
p
r
o
a
c
h
b
a
se
d
o
n
c
o
r
n
l
e
a
f
i
ma
g
e
c
l
a
ss
i
f
i
c
a
t
i
o
n
”
I
AES
I
n
t
e
rn
a
t
i
o
n
a
l
J
o
u
rn
a
l
o
f
Art
i
f
i
c
i
a
l
I
n
t
e
l
l
i
g
e
n
c
e
(
I
J
AI
)
,
v
o
l
.
1
3
,
n
o
.
1
,
2
0
2
4
,
p
p
.
1
1
3
9
-
1
1
4
8
,
d
o
i
:
1
0
.
1
1
5
9
1
/
i
j
a
i
.
v
1
3
.
i
1
.
p
p
1
1
3
9
-
1
1
4
8
.
[
8
]
S
.
H
.
S
r
e
e
d
h
a
r
a
,
V.
K
u
mar,
a
n
d
S
.
S
a
l
ma
,
“
Ef
f
i
c
i
e
n
t
b
i
g
d
a
t
a
c
l
u
s
t
e
r
i
n
g
u
si
n
g
a
d
h
o
c
f
u
z
z
y
c
me
a
n
s
a
n
d
a
u
t
o
-
e
n
c
o
d
e
r
C
N
N
,
”
i
n
I
n
v
e
n
t
i
v
e
C
o
m
p
u
t
a
t
i
o
n
a
n
d
I
n
f
o
rm
a
t
i
o
n
T
e
c
h
n
o
l
o
g
i
e
s
,
v
o
l
5
6
3
,
pp
.
353
–
36
,
2
0
2
3
,
d
o
i
:
1
0
.
1
0
0
7
/
9
7
8
-
9
8
1
-
19
-
7
4
0
2
-
1
_
2
5
.
[
9
]
C
.
R
a
o
a
n
d
X
.
Lo
u
,
“
M
u
l
t
i
p
l
i
c
a
t
i
o
n
a
n
d
a
c
c
u
m
u
l
a
t
i
o
n
c
o
-
o
p
t
i
mi
z
a
t
i
o
n
f
o
r
l
o
w
c
o
m
p
l
e
x
i
t
y
F
I
R
f
i
l
t
e
r
i
mp
l
e
m
e
n
t
a
t
i
o
n
,
”
El
e
c
t
r
o
n
i
c
s
,
v
o
l
.
1
1
,
n
o
.
1
1
,
2
0
2
2
,
d
o
i
:
1
0
.
3
3
9
0
/
e
l
e
c
t
r
o
n
i
c
s
1
1
1
1
1
7
2
1
.
[
1
0
]
N
.
P
a
n
,
Y
.
W
a
n
g
,
J.
C
h
e
n
,
a
n
d
J.
B
e
n
e
st
y
,
“
A
si
n
g
l
e
-
i
n
p
u
t
/
b
i
n
a
u
r
a
l
-
o
u
t
p
u
t
a
n
t
i
p
h
a
si
c
s
p
e
e
c
h
e
n
h
a
n
c
e
me
n
t
m
e
t
h
o
d
f
o
r
sp
e
e
c
h
i
n
t
e
l
l
i
g
i
b
i
l
i
t
y
i
mp
r
o
v
e
m
e
n
t
,
”
I
EEE
S
i
g
n
a
l
Pr
o
c
e
ssi
n
g
L
e
t
t
e
rs
,
v
o
l
.
2
8
,
p
p
.
1
4
4
5
–
1
4
4
9
,
2
0
2
1
,
d
o
i
:
1
0
.
1
1
0
9
/
LSP
.
2
0
2
1
.
3
0
9
5
0
1
6
.
[
1
1
]
A
.
P
r
o
d
e
u
s,
V
.
D
i
d
k
o
v
s
k
y
i
,
M
.
D
i
d
k
o
v
s
k
a
,
I
.
K
o
t
v
y
t
s
k
y
i
,
D
.
M
o
t
o
r
n
i
u
k
,
a
n
d
A
.
K
h
r
a
p
a
c
h
e
v
s
k
y
i
,
“
O
b
j
e
c
t
i
v
e
a
n
d
s
u
b
j
e
c
t
i
v
e
a
ssessm
e
n
t
o
f
t
h
e
q
u
a
l
i
t
y
a
n
d
i
n
t
e
l
l
i
g
i
b
i
l
i
t
y
o
f
n
o
i
s
e
d
sp
e
e
c
h
,
”
2
0
1
8
I
n
t
e
r
n
a
t
i
o
n
a
l
S
c
i
e
n
t
i
f
i
c
-
Pr
a
c
t
i
c
a
l
C
o
n
f
e
re
n
c
e
o
n
Pr
o
b
l
e
m
s
o
f
I
n
f
o
c
o
m
m
u
n
i
c
a
t
i
o
n
s
S
c
i
e
n
c
e
a
n
d
T
e
c
h
n
o
l
o
g
y
,
p
p
.
7
1
–
7
4
,
2
0
1
8
,
d
o
i
:
1
0
.
1
1
0
9
/
I
N
F
O
C
O
M
M
S
T.
2
0
1
8
.
8
6
3
2
1
2
5
.
[
1
2
]
S
.
P
e
z
z
u
l
l
i
,
M
.
G
.
M
a
r
t
i
n
i
,
a
n
d
N
.
B
a
r
man
,
“
Est
i
ma
t
i
o
n
o
f
q
u
a
l
i
t
y
s
c
o
r
e
s
f
r
o
m
s
u
b
j
e
c
t
i
v
e
t
e
s
t
s
-
b
e
y
o
n
d
su
b
j
e
c
t
s’
M
O
S
,
”
I
E
EE
T
ra
n
s
a
c
t
i
o
n
s
o
n
M
u
l
t
i
m
e
d
i
a
,
v
o
l
.
2
3
,
p
p
.
2
5
0
5
–
2
5
1
9
,
2
0
2
1
,
d
o
i
:
1
0
.
1
1
0
9
/
T
M
M
.
2
0
2
0
.
3
0
1
3
3
4
9
.
[
1
3
]
R
.
S
.
P
u
j
a
r
a
n
d
P
.
N
.
K
u
l
k
a
r
n
i
,
“
C
a
sca
d
e
d
s
t
r
u
c
t
u
r
e
o
f
w
i
e
n
e
r
f
i
l
t
e
r
w
i
t
h
F
B
S
b
a
s
e
d
sp
e
c
t
r
a
l
s
p
l
i
t
t
i
n
g
a
n
d
d
y
n
a
mi
c
r
a
n
g
e
c
o
m
p
r
e
ss
i
o
n
f
o
r
l
i
st
e
n
e
r
s
w
i
t
h
se
n
so
r
i
n
e
u
r
a
l
h
e
a
r
i
n
g
l
o
ss,
”
I
N
D
I
C
O
N
2
0
1
8
-
1
5
t
h
I
EEE
I
n
d
i
a
C
o
u
n
c
i
l
I
n
t
e
r
n
a
t
i
o
n
a
l
C
o
n
f
e
re
n
c
e
,
2
0
1
8
,
d
o
i
:
1
0
.
1
1
0
9
/
I
N
D
I
C
O
N
4
5
5
9
4
.
2
0
1
8
.
8
9
8
7
0
5
8
.
[
1
4
]
Y
.
L
e
n
g
,
X
.
T
a
n
,
S
.
Z
h
a
o
,
F
.
S
o
o
n
g
,
X
.
Y
.
L
i
,
a
n
d
T
.
Q
i
n
,
“
M
B
N
e
t
:
M
O
S
p
r
e
d
i
c
t
i
o
n
f
o
r
s
y
n
t
h
e
s
i
z
e
d
s
p
e
e
c
h
w
i
t
h
m
e
a
n
-
b
i
a
s
n
e
t
w
o
r
k
,
”
I
E
E
E
I
n
t
e
r
n
a
t
i
o
n
a
l
C
o
n
f
e
r
e
n
c
e
o
n
A
c
o
u
s
t
i
c
s
,
S
p
e
e
c
h
a
n
d
S
i
g
n
a
l
P
r
o
c
e
s
s
i
n
g
,
p
p
.
3
9
1
–
3
9
5
,
2
0
2
1
,
d
o
i
:
1
0
.
1
1
0
9
/
I
C
A
S
S
P
3
9
7
2
8
.
2
0
2
1
.
9
4
1
3
8
7
7
.
[
1
5
]
S
.
S
a
n
g
e
e
t
h
a
a
n
d
P
.
K
a
n
n
a
n
,
“
D
e
s
i
g
n
a
n
d
a
n
a
l
y
s
i
s
o
f
d
i
g
i
t
a
l
f
i
l
t
e
r
s
f
o
r
s
p
e
e
c
h
si
g
n
a
l
u
si
n
g
m
u
l
t
i
r
a
t
e
si
g
n
a
l
p
r
o
c
e
ssi
n
g
,
”
I
C
T
A
C
T
J
o
u
rn
a
l
o
n
M
i
c
r
o
e
l
e
c
t
r
o
n
i
c
s
,
v
o
l
.
3
,
n
o
.
4
,
p
p
.
4
8
0
–
4
8
7
,
2
0
1
8
.
[
1
6
]
P
.
D
e
r
l
e
t
h
e
t
a
l
.
,
“
B
i
n
a
u
r
a
l
si
g
n
a
l
p
r
o
c
e
ss
i
n
g
i
n
h
e
a
r
i
n
g
a
i
d
s
,
”
S
e
m
i
n
a
rs
i
n
H
e
a
ri
n
g
,
v
o
l
.
4
2
,
n
o
.
3
,
p
p
.
2
0
6
–
2
2
3
,
2
0
2
1
,
d
o
i
:
1
0
.
1
0
5
5
/
s
-
0
0
4
1
-
1
7
3
5
1
7
6
.
[
1
7
]
P
.
P
r
a
k
a
sh
,
A
.
S
r
e
e
d
h
a
r
,
J.
R
.
B
a
l
a
n
,
a
n
d
A
.
M
.
V
a
r
g
h
e
s
e
,
“
B
e
n
e
f
i
t
o
n
d
a
i
l
y
l
i
st
e
n
i
n
g
w
i
t
h
t
e
c
h
n
o
l
o
g
i
c
a
l
a
d
v
a
n
c
e
me
n
t
s
:
c
o
m
p
a
r
i
so
n
o
f
b
a
s
i
c
a
n
d
p
r
e
m
i
u
m
c
a
t
e
g
o
r
y
h
e
a
r
i
n
g
a
i
d
s,”
Eu
r
o
p
e
a
n
Arc
h
i
v
e
s
o
f
O
t
o
-
R
h
i
n
o
-
L
a
ry
n
g
o
l
o
g
y
,
v
o
l
.
2
7
9
,
n
o
.
6
,
p
p
.
3
1
7
9
–
3
1
8
7
,
2
0
2
2
,
d
o
i
:
1
0
.
1
0
0
7
/
s
0
0
4
0
5
-
0
2
1
-
0
7
2
4
0
-
3.
[
1
8
]
J.
A
mi
n
i
,
R
.
C
.
H
e
n
d
r
i
k
s
,
R
.
H
e
u
sd
e
n
s,
M
.
G
u
o
,
a
n
d
J
.
Je
n
s
e
n
,
“
S
p
a
t
i
a
l
l
y
c
o
r
r
e
c
t
r
a
t
e
-
c
o
n
s
t
r
a
i
n
e
d
n
o
i
s
e
r
e
d
u
c
t
i
o
n
f
o
r
b
i
n
a
u
r
a
l
h
e
a
r
i
n
g
a
i
d
s
i
n
w
i
r
e
l
e
ss
a
c
o
u
st
i
c
s
e
n
s
o
r
n
e
t
w
o
r
k
s,”
I
EEE
/
A
C
M
T
ra
n
s
a
c
t
i
o
n
s
o
n
A
u
d
i
o
S
p
e
e
c
h
a
n
d
L
a
n
g
u
a
g
e
Pr
o
c
e
ssi
n
g
,
v
o
l
.
2
8
,
p
p
.
2
7
3
1
–
2
7
4
2
,
2
0
2
0
,
d
o
i
:
1
0
.
1
1
0
9
/
T
A
S
LP.2
0
2
0
.
3
0
2
8
2
6
4
.
[
1
9
]
B
.
S
.
L
.
C
h
a
i
t
a
n
y
a
a
n
d
K
.
R
a
j
a
l
a
k
s
h
mi
,
“
F
i
r
a
n
d
f
o
l
d
e
d
i
i
r
f
i
l
t
e
r
d
e
si
g
n
s
f
o
r
sp
e
e
c
h
p
r
o
c
e
ss
i
n
g
i
n
h
e
a
r
i
n
g
a
i
d
,
”
I
n
t
e
r
n
a
t
i
o
n
a
l
J
o
u
rn
a
l
o
f
N
o
n
l
i
n
e
a
r
An
a
l
y
si
s
a
n
d
A
p
p
l
i
c
a
t
i
o
n
s
,
v
o
l
.
1
2
,
p
p
.
1
6
3
1
–
1
6
3
8
,
2
0
2
1
,
d
o
i
:
1
0
.
2
2
0
7
5
/
I
JN
A
A
.
2
0
2
1
.
5
8
4
9
.
[
2
0
]
G
.
Y
a
o
,
L
.
Li
,
W
.
L
v
,
a
n
d
H
.
Zh
a
n
g
,
“
A
d
e
g
e
n
e
r
a
t
e
s
p
a
t
i
a
l
A
R
M
A
p
r
o
c
e
s
s
o
f
e
x
t
e
r
n
a
l
s
o
u
r
c
e
s
i
mp
i
n
g
i
n
g
u
p
o
n
a
n
a
z
i
m
u
t
h
-
o
n
l
y
U
LA
a
n
d
t
h
e
e
s
t
i
m
a
t
i
o
n
o
f
d
o
e
s
a
n
d
n
o
i
se
v
a
r
i
a
n
c
e
s
,
”
I
EE
E
Ac
c
e
ss
,
v
o
l
.
8
,
p
p
.
2
2
1
8
7
1
–
2
2
1
8
8
5
,
2
0
2
0
,
d
o
i
:
1
0
.
1
1
0
9
/
A
C
C
ESS
.
2
0
2
0
.
3
0
4
1
0
1
2
.
[
2
1
]
C
.
T
.
C
h
e
n
,
L.
M
.
C
h
a
n
g
,
a
n
d
C
.
H
.
Lo
h
,
“
A
r
e
v
i
e
w
o
f
sp
e
c
t
r
a
l
a
n
a
l
y
s
i
s
f
o
r
l
o
w
-
f
r
e
q
u
e
n
c
y
t
r
a
n
s
i
e
n
t
v
i
b
r
a
t
i
o
n
s,
”
J
o
u
r
n
a
l
o
f
L
o
w
Fre
q
u
e
n
c
y
N
o
i
se
Vi
b
r
a
t
i
o
n
a
n
d
A
c
t
i
v
e
C
o
n
t
r
o
l
,
v
o
l
.
4
0
,
n
o
.
2
,
p
p
.
6
5
6
–
6
7
1
,
2
0
2
1
,
d
o
i
:
1
0
.
1
1
7
7
/
1
4
6
1
3
4
8
4
2
0
9
4
2
0
0
8
.
[
2
2
]
P
.
O
k
o
n
i
e
w
s
k
i
a
n
d
J
.
P
i
s
k
o
r
o
w
s
k
i
,
“
A
c
o
n
c
e
p
t
o
f
I
I
R
f
i
l
t
e
r
s
w
i
t
h
t
i
m
e
-
v
a
r
y
i
n
g
c
o
e
f
f
i
c
i
e
n
t
s
a
n
d
e
q
u
a
l
i
se
d
g
r
o
u
p
d
e
l
a
y
r
e
s
p
o
n
s
e
,
”
Me
a
su
r
e
m
e
n
t
,
v
o
l
.
6
0
,
p
p
.
1
3
–
2
4
,
2
0
1
5
,
d
o
i
:
1
0
.
1
0
1
6
/
j
.
mea
su
r
e
me
n
t
.
2
0
1
4
.
0
9
.
0
7
7
.
[
2
3
]
W
.
Y
a
o
,
Y
.
Y
a
n
g
,
Y
.
X
u
,
F
.
B
l
a
a
b
j
e
r
g
,
S
.
Li
u
,
a
n
d
G
.
W
i
l
s
o
n
,
“
P
h
a
se
r
e
s
h
a
p
i
n
g
v
i
a
a
l
l
-
p
a
ss
f
i
l
t
e
r
s
f
o
r
r
o
b
u
st
LC
L
-
f
i
l
t
e
r
a
c
t
i
v
e
d
a
m
p
i
n
g
,
”
I
EEE
T
ra
n
s
a
c
t
i
o
n
s
o
n
Po
w
e
r
El
e
c
t
r
o
n
i
c
s
,
v
o
l
.
3
5
,
n
o
.
3
,
p
p
.
3
1
1
4
–
3
1
2
6
,
2
0
2
0
,
d
o
i
:
1
0
.
1
1
0
9
/
TP
EL.
2
0
1
9
.
2
9
2
7
2
7
2
.
[
2
4
]
A
.
W
.
R
i
x
,
J.
G
.
B
e
e
r
e
n
d
s
,
M
.
P
.
H
o
l
l
i
e
r
,
a
n
d
A
.
P
.
H
e
k
st
r
a
,
“
P
e
r
c
e
p
t
u
a
l
e
v
a
l
u
a
t
i
o
n
o
f
sp
e
e
c
h
q
u
a
l
i
t
y
(
P
ESQ
)
-
A
n
e
w
me
t
h
o
d
f
o
r
sp
e
e
c
h
q
u
a
l
i
t
y
a
ss
e
ssm
e
n
t
o
f
t
e
l
e
p
h
o
n
e
n
e
t
w
o
r
k
s
a
n
d
c
o
d
e
c
s,
”
I
C
A
S
S
P,
I
E
EE
I
n
t
e
rn
a
t
i
o
n
a
l
C
o
n
f
e
re
n
c
e
o
n
A
c
o
u
st
i
c
s
,
S
p
e
e
c
h
a
n
d
S
i
g
n
a
l
Pr
o
c
e
ssi
n
g
-
Pr
o
c
e
e
d
i
n
g
s
,
v
o
l
.
2
,
p
p
.
7
4
9
–
7
5
2
,
2
0
0
1
,
d
o
i
:
1
0
.
1
1
0
9
/
i
c
a
ssp
.
2
0
0
1
.
9
4
1
0
2
3
.
[
2
5
]
I
TU
-
T,
“
M
e
t
h
o
d
s f
o
r
su
b
j
e
c
t
i
v
e
d
e
t
e
r
mi
n
a
t
i
o
n
o
f
t
r
a
n
s
mi
ss
i
o
n
q
u
a
l
i
t
y
,
”
I
n
t
e
rn
a
t
i
o
n
a
l
T
e
l
e
c
o
m
m
u
n
i
c
a
t
i
o
n
U
n
i
o
n
,
p
p
.
1
-
2
8
,
1
9
9
6
.
B
I
O
G
RAP
H
I
E
S O
F
AUTH
O
RS
Apa
rna
Chila
k
a
wa
d
co
m
p
leted
h
er
B
.
E
.
d
eg
r
e
e
f
r
o
m
Kar
n
ata
k
a
Un
iv
er
s
ity
,
Dh
ar
wad
an
d
M
.
T
ec
h
.
d
eg
r
ee
in
Dig
ital
E
lectr
o
n
ics
f
r
o
m
Vis
v
esv
ar
ay
a
T
ec
h
n
o
lo
g
ical
Un
iv
er
s
ity
,
B
elg
au
m
.
Kar
n
atak
a,
I
n
d
ia.
Sh
e
is
p
u
r
s
u
in
g
h
e
r
Ph
.
D.
in
Dep
ar
tm
en
t
o
f
E
C
E
,
B
asav
esh
war
E
n
g
in
ee
r
in
g
C
o
lleg
e
(
Au
to
n
o
m
o
u
s
)
,
B
ag
alk
o
te,
Vis
v
esv
ar
ay
a
T
ec
h
n
o
lo
g
ical
Un
iv
er
s
ity
,
B
elag
av
i
.
Kar
n
atak
a,
I
n
d
ia.
Her
r
esear
c
h
in
ter
ests
ar
e
s
p
ee
ch
p
r
o
ce
s
s
in
g
an
d
d
ig
ital
s
ig
n
al
p
r
o
ce
s
s
in
g
.
Sh
e
ca
n
b
e
co
n
tacte
d
a
t
em
ail: a
p
ar
n
a
c
_
1
2
@
r
ed
if
f
m
ail
.
co
m
.
Dr.
P
a
nd
ura
ng
a
ra
o
N.
K
ulk
a
rni
h
as
co
m
p
leted
h
is
Ph
.
D
.
f
r
o
m
I
n
d
ia
n
I
n
s
titu
te
o
f
T
ec
h
n
o
lo
g
y
B
o
m
b
ay
,
I
n
d
ia,
in
th
e
y
ea
r
2
0
1
0
.
He
h
as
th
ir
ty
-
f
o
u
r
y
ea
r
s
o
f
teac
h
in
g
e
x
p
er
ien
ce
an
d
f
o
u
r
teen
y
ea
r
s
o
f
r
esear
ch
ex
p
er
ien
ce
.
C
u
r
r
en
tly
h
e
is
wo
r
k
in
g
as
ac
ad
em
ic
Dea
n
an
d
Pro
f
ess
o
r
,
Dep
ar
tm
en
t
o
f
E
lectr
o
n
ics
an
d
C
o
m
m
u
n
icatio
n
E
n
g
in
ee
r
in
g
,
at
B
asav
esh
war
E
n
g
in
ee
r
in
g
C
o
lleg
e,
B
ag
alk
o
te
.
He
ca
n
b
e
co
n
tacte
d
at
em
ail: p
n
k
_
b
ewo
o
r
@
y
ah
o
o
.
c
o
m
.
Evaluation Warning : The document was created with Spire.PDF for Python.