I
nte
rna
t
io
na
l J
o
urna
l o
f
E
lect
rica
l a
nd
Co
m
p
ute
r
E
ng
in
ee
ring
(
I
J
E
CE
)
Vo
l.
7
,
No
.
2
,
A
p
r
il
201
7
,
p
p
.
84
2
~
8
49
I
SS
N:
2
0
8
8
-
8708
,
DOI
: 1
0
.
1
1
5
9
1
/
i
j
ec
e
.
v7
i
2
.
p
p
8
4
2
-
8
4
9
842
J
o
ur
na
l ho
m
ep
a
g
e
:
h
ttp
:
//ia
e
s
jo
u
r
n
a
l.c
o
m/o
n
lin
e/in
d
ex
.
p
h
p
/I
JE
C
E
Fu
zz
y
Re
cursiv
e
Lea
st
-
Squa
res
Ap
pro
a
ch in
Spe
ech
Sy
ste
m
Identifica
tion: A
Tra
nsfo
r
m
e
d Do
m
a
in L
PC
M
o
del
K
a
h Wa
i C
hea
h
,
No
o
r
At
ina
h Ah
m
a
d
S
c
h
o
o
l
o
f
M
a
th
e
m
a
ti
c
a
l
S
c
ien
c
e
s,
Un
iv
e
rsiti
S
a
in
s
M
a
lay
sia
,
M
a
la
y
sia
Art
icle
I
nfo
AB
ST
RAC
T
A
r
ticle
his
to
r
y:
R
ec
eiv
ed
Dec
27
,
2
0
1
6
R
ev
i
s
ed
Feb
25
,
2
0
1
7
A
cc
ep
ted
Mar
14
,
2
0
1
7
In
sp
e
e
c
h
sy
ste
m
id
e
n
ti
f
ica
ti
o
n
,
li
n
e
a
r
p
re
d
ictiv
e
c
o
d
i
n
g
(L
P
C)
m
o
d
e
l
is
o
f
ten
e
m
p
lo
y
e
d
d
u
e
to
it
s
sim
p
le
y
e
t
p
o
w
e
r
f
u
l
re
p
re
se
n
tatio
n
o
f
sp
e
e
c
h
p
ro
d
u
c
ti
o
n
m
o
d
e
l.
Ho
w
e
v
e
r,
th
e
a
c
c
u
ra
c
y
o
f
L
P
C
m
o
d
e
l
o
f
ten
d
e
p
e
n
d
s
o
n
th
e
n
u
m
b
e
r
a
n
d
q
u
a
li
ty
o
f
p
a
st
s
p
e
e
c
h
sa
m
p
les
th
a
t
a
re
f
e
d
in
to
t
h
e
m
o
d
e
l;
a
n
d
it
b
e
c
o
m
e
s
a
p
ro
b
lem
w
h
e
n
p
a
st
sp
e
e
c
h
s
a
m
p
les
a
re
n
o
t
w
id
e
l
y
a
v
a
il
a
b
le
o
r
c
o
rru
p
ted
b
y
n
o
ise
.
In
th
is
p
a
p
e
r,
f
u
z
z
y
s
y
ste
m
is
in
teg
ra
t
e
d
in
to
th
e
L
P
C
m
o
d
e
l
u
sin
g
th
e
re
c
u
rsiv
e
lea
st
-
sq
u
a
re
s
a
p
p
ro
a
c
h
,
w
h
e
re
th
e
f
u
z
z
y
p
a
ra
m
e
ters
a
re
u
se
d
to
c
h
a
ra
c
teriz
e
th
e
g
i
v
e
n
sp
e
e
c
h
s
a
m
p
les
.
T
h
is
tran
sf
o
r
m
e
d
d
o
m
a
in
L
P
C
m
o
d
e
l
i
s
c
a
ll
e
d
th
e
F
RL
S
-
L
P
C
m
o
d
e
l,
in
w
h
ich
it
s
p
e
rf
o
r
m
a
n
c
e
d
e
p
e
n
d
s
o
n
th
e
f
u
z
z
y
ru
les
a
n
d
m
e
m
b
e
rsh
ip
f
u
n
c
ti
o
n
s
d
e
f
in
e
d
b
y
th
e
u
se
r.
Ba
se
d
o
n
t
h
e
sim
u
latio
n
s,
th
e
F
RL
S
-
L
P
C
m
o
d
e
l
w
it
h
t
h
is
sp
e
c
ial
p
ro
p
e
rty
is
sh
o
w
n
to
o
u
t
p
e
rf
o
rm
th
e
L
P
C
m
o
d
e
l.
Un
d
e
r
th
e
c
o
n
d
it
io
n
o
f
li
m
it
e
d
p
a
st
sp
e
e
c
h
sa
m
p
les
,
sim
u
latio
n
re
su
lt
sh
o
w
s
th
a
t
th
e
s
y
n
th
e
ti
c
sp
e
e
c
h
p
ro
d
u
c
e
d
b
y
th
e
F
R
L
S
-
LP
C
m
o
d
e
l
is
b
e
tt
e
r
th
a
n
th
o
se
p
r
o
d
u
c
e
d
b
y
th
e
L
P
C
m
o
d
e
l
in
term
s
o
f
p
re
d
ictio
n
e
rro
r.
F
u
rt
h
e
rm
o
re
w
it
h
c
o
rru
p
ted
p
a
st
sp
e
e
c
h
sa
m
p
les
,
th
e
F
RL
S
-
L
P
C
m
o
d
e
l
is
a
b
le
to
p
r
o
v
id
e
b
e
tt
e
r
re
c
o
n
stru
c
ted
sp
e
e
c
h
w
h
il
e
th
e
L
P
C
m
o
d
e
l
is
f
a
il
e
d
to
d
o
so
.
K
ey
w
o
r
d
:
Fu
zz
y
r
ec
u
r
s
i
v
e
l
ea
s
t
-
s
q
u
ar
es
L
i
n
ea
r
p
r
ed
ictiv
e
c
o
d
in
g
Sp
ee
ch
s
y
s
te
m
i
d
en
t
if
icatio
n
Co
p
y
rig
h
t
©
2
0
1
7
In
stit
u
te o
f
A
d
v
a
n
c
e
d
E
n
g
i
n
e
e
rin
g
a
n
d
S
c
ien
c
e
.
Al
l
rig
h
ts
re
se
rv
e
d
.
C
o
r
r
e
s
p
o
nd
ing
A
uth
o
r
:
Kah
W
ai
C
h
ea
h
,
Sch
o
o
l o
f
Ma
t
h
e
m
at
ical
Scie
n
ce
s
,
Un
i
v
er
s
iti
Sain
s
Ma
la
y
s
ia,
1
1
8
0
0
USM,
P
u
lau
P
in
an
g
,
Ma
la
y
s
ia.
E
m
ail:
c
k
w
1
3
_
m
ah
0
2
8
@
s
tu
d
en
t.u
s
m
.
m
y
1.
I
NT
RO
D
UCT
I
O
N
I
n
s
p
ee
ch
p
r
o
ce
s
s
in
g
,
th
e
p
ar
a
m
eter
izatio
n
o
f
an
an
a
lo
g
s
p
ee
ch
s
i
g
n
al
i
s
an
i
m
p
o
r
tan
t
s
t
ep
,
as
th
e
r
esu
lted
p
ar
a
m
e
ter
s
s
h
o
u
ld
r
ep
r
esen
t
t
h
e
s
alie
n
t
s
p
ec
tr
al
en
er
g
ies
o
f
t
h
e
s
o
u
n
d
[
1
]
.
Sin
ce
th
e
li
n
ea
r
p
r
ed
ictiv
e
co
d
in
g
(
L
P
C
)
m
o
d
el
p
r
o
v
id
es
a
g
o
o
d
ap
p
r
o
x
i
m
a
tio
n
to
th
e
v
o
ca
l
tr
ac
t
s
p
ec
tr
al
en
v
elo
p
e
in
s
u
c
h
a
w
a
y
th
at
th
e
p
ar
s
i
m
o
n
io
u
s
r
ep
r
esen
tat
i
o
n
o
f
t
h
e
v
o
ca
l
tr
ac
t
c
h
ar
ac
ter
is
tics
b
ec
o
m
e
s
p
o
s
s
ib
le
[
2
]
,
th
e
L
P
C
m
o
d
el
i
s
th
e
m
o
s
t
co
m
m
o
n
m
o
d
el
u
s
ed
i
n
s
p
ee
c
h
s
p
ec
tr
al
a
n
al
y
s
i
s
.
B
y
c
h
a
n
g
in
g
s
p
ec
tr
al
an
al
y
s
i
s
in
w
a
v
ef
o
r
m
d
ata
in
ter
v
a
l
to
s
p
ec
tr
o
g
r
ap
h
ic
ti
m
e
-
f
r
eq
u
en
c
y
d
o
m
ai
n
w
h
er
e
th
e
in
f
o
r
m
atio
n
(
s
u
c
h
as
in
ter
-
f
o
r
m
a
n
t
e
n
er
g
y
f
ill)
ca
n
s
i
g
n
i
f
ica
n
tl
y
b
e
p
o
r
tr
ay
ed
[
3
]
,
th
e
co
ef
f
icie
n
ts
o
f
L
P
C
m
o
d
el
p
r
o
v
e
its
co
n
tr
ib
u
tio
n
s
in
th
e
ap
p
licatio
n
o
f
s
p
ee
ch
s
i
g
n
als
s
y
n
t
h
es
i
s
[
4
]
.
I
n
m
o
s
t
r
ec
en
t
d
ec
ad
e,
th
e
L
P
C
m
o
d
el
h
as
b
ee
n
i
m
p
l
e
m
en
ted
in
v
ar
io
u
s
ap
p
licatio
n
s
s
u
c
h
as
lo
n
g
te
r
m
r
ec
o
r
d
in
g
s
o
f
elec
tr
o
m
y
o
g
r
ap
h
y
s
ig
n
al
s
[
5
]
,
r
ec
o
g
n
iti
o
n
o
f
Ma
la
y
ala
m
v
o
w
el
[
6
]
,
clu
s
ter
i
n
g
o
f
m
icr
o
ar
r
ay
g
e
n
etic
d
ata
[
7
]
,
r
ec
o
n
s
t
r
u
ctio
n
o
f
m
i
s
s
i
n
g
elec
tr
o
ca
r
d
io
g
r
a
m
s
i
g
n
a
ls
[
8
]
,
d
y
n
a
m
ic
te
x
t
u
r
e
s
e
g
m
e
n
tatio
n
o
f
i
m
a
g
e
s
eq
u
e
n
ce
s
[
9
]
,
an
d
class
i
f
icatio
n
o
f
h
u
m
a
n
ac
ti
v
it
y
b
ased
o
n
m
icr
o
-
d
o
p
p
ler
s
ig
n
at
u
r
es
[
1
0
]
.
A
ll
o
f
th
ese
ap
p
licatio
n
s
h
a
v
e
p
r
o
v
en
th
at
th
e
L
P
C
m
o
d
el
co
u
ld
b
e
im
p
le
m
e
n
ted
in
a
m
o
r
e
g
en
er
a
l a
p
p
r
o
ac
h
.
Ho
w
e
v
er
in
s
p
ee
ch
p
r
o
ce
s
s
in
g
,
th
e
ac
cu
r
ac
y
o
f
L
P
C
m
o
d
el
o
f
te
n
d
ep
en
d
s
o
n
th
e
n
u
m
b
er
an
d
q
u
alit
y
o
f
p
ast
s
p
ee
ch
s
a
m
p
les
t
h
at
ar
e
f
ed
in
to
th
e
m
o
d
el.
Stu
d
y
s
h
o
w
s
th
at,
t
h
e
r
ea
s
o
n
ab
le
n
u
m
b
er
o
f
p
ast
s
p
ee
c
h
s
a
m
p
les
th
at
i
s
f
ed
i
n
to
th
e
L
P
C
m
o
d
el
to
ap
p
r
o
x
i
m
ate
cu
r
r
e
n
t sp
ee
ch
s
a
m
p
le
d
ep
en
d
s
o
n
t
h
e
s
a
m
p
lin
g
r
ate
o
f
Evaluation Warning : The document was created with Spire.PDF for Python.
I
J
E
C
E
I
SS
N:
2
0
8
8
-
8708
F
u
z
z
y
R
ec
u
r
s
ive
Lea
s
t
-
S
q
u
a
r
es A
p
p
r
o
a
ch
in
S
p
ee
c
h
S
yst
em
I
d
en
tifi
ca
tio
n
.
.
.
.
(
K
a
h
W
a
i Ch
ea
h
)
843
th
e
g
i
v
en
s
p
ee
ch
.
Fo
r
a
s
p
ee
ch
w
it
h
s
a
m
p
lin
g
r
ate
o
f
,
th
e
n
u
m
b
er
o
f
p
ast
s
p
ee
ch
s
a
m
p
les
in
v
o
l
v
ed
is
w
i
th
t
h
e
ad
d
itio
n
o
f
t
w
o
to
f
i
v
e
p
ast
s
p
ee
ch
s
a
m
p
les
[
1
1
]
.
T
y
p
icall
y
,
f
o
r
s
p
ee
ch
s
a
m
p
le,
th
e
n
u
m
b
er
o
f
p
ast
s
p
ee
ch
s
a
m
p
le
s
,
is
eq
u
al
to
1
0
,
w
h
er
e
th
e
ad
d
itio
n
al
t
w
o
p
ast
s
p
ee
ch
s
a
m
p
les
ar
e
u
s
ed
to
f
o
r
m
u
la
te
t
h
e
g
lo
tt
al
f
lo
w
a
n
d
th
e
r
ad
iatio
n
ad
d
ed
d
u
r
in
g
th
e
p
r
o
n
u
n
c
iatio
n
o
f
s
p
ee
ch
[
2
]
.
I
f
is
as
lo
w
as
t
w
o
,
al
t
h
o
u
g
h
t
h
e
s
y
n
t
h
etic
s
p
ee
ch
is
s
till
i
n
telli
g
ib
le,
y
et
it
i
s
p
o
o
r
in
q
u
alit
y
[
4
]
.
T
h
is
s
h
o
w
s
t
h
at,
in
o
r
d
er
to
h
av
e
a
h
i
g
h
q
u
a
lit
y
o
f
s
y
n
t
h
etic
s
p
ee
c
h
,
th
e
n
u
m
b
er
o
f
p
ast s
p
ee
ch
s
a
m
p
les t
h
at
is
f
ed
i
n
to
th
e
L
P
C
m
o
d
el
m
u
s
t
b
e
s
u
f
f
icie
n
tl
y
lar
g
e
s
o
th
at
t
h
e
co
ef
f
icie
n
ts
p
r
o
d
u
ce
d
ar
e
ab
le
to
ch
ar
ac
ter
ize
th
e
s
alie
n
t
s
p
ec
tr
al
en
er
g
ie
s
o
f
t
h
e
s
o
u
n
d
.
Fro
m
t
h
e
f
o
r
m
u
latio
n
o
f
L
P
C
m
o
d
el,
p
ast
s
p
ee
c
h
s
a
m
p
les
f
o
r
m
a
li
n
ea
r
co
m
b
in
at
io
n
w
it
h
t
h
e
L
P
C
co
ef
f
icie
n
t
s
to
ap
p
r
o
x
i
m
ate
c
u
r
r
en
t
s
p
ee
ch
s
a
m
p
le.
Fo
r
s
u
c
h
,
th
e
r
esu
lt
ed
L
P
C
co
ef
f
icie
n
ts
ai
m
to
m
i
n
i
m
ize
th
e
p
r
ed
ictio
n
er
r
o
r
at
ev
er
y
t
i
m
e
i
n
s
ta
n
t
.
I
n
liter
atu
r
e,
f
u
z
z
y
s
y
s
te
m
is
o
f
ten
e
m
p
lo
y
ed
to
g
eth
er
w
it
h
t
h
e
ap
p
licatio
n
s
o
f
L
P
C
m
o
d
el
to
ac
h
iev
e
ce
r
tai
n
o
b
j
ec
tiv
es.
Fo
r
ex
a
m
p
le,
i
n
v
o
ice
o
v
er
in
ter
n
et
p
r
o
to
co
l
s
y
s
te
m
,
ev
o
lv
i
n
g
T
ak
ag
i
-
S
u
g
en
o
f
u
zz
y
m
o
d
el
is
u
s
ed
to
r
ec
o
v
er
t
h
e
m
i
s
s
i
n
g
li
n
ea
r
s
p
ec
tr
al
p
air
s
t
h
at
ca
lc
u
lated
f
r
o
m
th
e
L
P
C
m
o
d
el
[
1
2
]
;
in
ca
n
ce
r
class
i
f
icatio
n
,
m
o
d
i
f
ied
f
u
zz
y
c
-
m
ea
n
s
al
g
o
r
ith
m
(
f
u
zz
y
cl
u
s
ter
i
n
g
)
is
u
s
ed
to
class
i
f
y
th
e
f
ea
t
u
r
es
t
h
at
ar
e
ex
tr
ac
ted
u
s
in
g
L
P
C
m
o
d
el
[
1
3
]
;
an
d
in
s
p
ee
ch
co
d
in
g
tr
an
s
m
is
s
io
n
,
f
u
zz
y
clu
s
ter
i
n
g
is
u
s
ed
to
clu
s
ter
v
o
iced
s
eg
m
e
n
t
w
h
er
e
t
h
ese
s
eg
m
e
n
ts
ar
e
to
b
e
tr
an
s
m
itte
d
to
g
eth
er
w
it
h
th
e
ex
tr
ac
ted
L
P
C
co
ef
f
icie
n
ts
[
1
4
]
.
No
tice
th
at,
f
u
zz
y
s
y
s
te
m
is
o
n
l
y
ap
p
lied
af
ter
w
a
r
d
s
an
d
th
e
L
P
C
co
ef
f
icie
n
t
s
ar
e
s
till
u
s
ed
in
t
h
ese
ap
p
licatio
n
s
.
I
n
t
h
is
p
ap
er
,
th
e
f
u
zz
y
s
y
s
t
e
m
i
s
i
n
te
g
r
ated
d
ir
ec
tl
y
in
to
th
e
L
P
C
m
o
d
el
u
s
in
g
r
ec
u
r
s
iv
e
least
-
s
q
u
ar
es
(
R
L
S)
ap
p
r
o
ac
h
.
I
n
s
t
ea
d
o
f
d
ir
ec
tly
f
ee
d
i
n
g
th
e
p
ast
s
p
ee
ch
s
a
m
p
les
i
n
to
th
e
L
P
C
m
o
d
el
an
d
s
o
lv
e
f
o
r
th
e
L
P
C
co
ef
f
icien
ts
,
w
e
tr
an
s
f
o
r
m
th
e
L
P
C
m
o
d
el
i
n
to
f
u
zz
y
d
o
m
ai
n
a
n
d
u
s
e
f
u
zz
y
p
ar
a
m
eter
s
to
ch
ar
ac
ter
ize
t
h
e
g
iv
e
n
s
p
ee
ch
s
a
m
p
les.
I
t
h
as
b
ee
n
p
r
o
v
en
t
h
at
t
h
e
f
u
zz
y
b
asis
f
u
n
ctio
n
[
1
5
]
an
d
its
r
ed
u
ce
d
f
o
r
m
[
1
6
]
ar
e
u
n
iv
er
s
a
l
ap
p
r
o
x
i
m
ato
r
.
Af
ter
a
b
r
ief
d
escr
ip
tio
n
o
f
th
e
f
o
r
m
u
latio
n
o
f
L
P
C
m
o
d
el
in
Sectio
n
2
,
th
e
tr
an
s
f
o
r
m
ed
d
o
m
ai
n
L
P
C
m
o
d
el
w
it
h
f
u
zz
y
r
ec
u
r
s
i
v
e
least
-
s
q
u
ar
es
ap
p
r
o
ac
h
(
F
R
L
S
-
L
P
C
)
w
ill
b
e
f
o
r
m
u
lated
in
Sect
io
n
3
.
I
n
Sectio
n
4
,
b
ased
o
n
th
e
co
n
f
i
g
u
r
atio
n
o
f
s
y
s
te
m
id
e
n
ti
f
icat
i
o
n
,
s
i
m
u
latio
n
s
ar
e
p
er
f
o
r
m
ed
o
n
t
h
e
r
ea
l
s
p
ee
ch
s
a
m
p
les
an
d
th
e
p
er
f
o
r
m
a
n
ce
o
f
F
R
L
S
-
L
P
C
m
o
d
el
is
e
v
al
u
ated
in
ter
m
s
o
f
t
h
e
p
r
ed
ictio
n
er
r
o
r
an
d
th
e
q
u
alit
y
o
f
s
y
n
th
et
ic
s
p
ee
ch
u
s
i
n
g
f
u
zz
y
p
ar
a
m
eter
s
.
Fin
al
l
y
,
co
n
clu
s
io
n
is
d
r
a
w
n
i
n
Sectio
n
5
.
2.
L
I
N
E
AR
P
R
E
DIC
T
I
VE
CO
DIN
G
Giv
e
n
a
s
p
ee
ch
s
a
m
p
le
at
cu
r
r
en
t
ti
m
e
in
s
tan
t
,
w
h
er
e
,
th
e
L
P
C
m
o
d
el
s
u
g
g
e
s
ts
t
h
at
t
h
e
c
u
r
r
en
t
s
p
ee
ch
s
a
m
p
le
ca
n
b
e
ap
p
r
o
x
i
m
a
ted
as
a
li
n
ea
r
co
m
b
i
n
atio
n
o
f
p
ast
s
p
ee
ch
s
a
m
p
les
s
u
c
h
t
h
at
,
(
1
)
w
h
er
e
ar
e
th
e
L
P
C
co
e
f
f
icien
ts
f
o
r
to
p
ast
s
p
ee
ch
s
a
m
p
le,
in
w
h
ic
h
ar
e
ass
u
m
ed
to
b
e
co
n
s
ta
n
t
o
v
er
th
e
s
p
ee
ch
an
al
y
s
i
s
f
r
a
m
e
w
it
h
s
p
ee
ch
s
a
m
p
les.
T
h
e
o
b
j
ec
tiv
e
f
u
n
c
tio
n
o
f
L
P
C
m
o
d
el
is
g
iv
e
n
b
y
∑
∑
∑
,
(
2
)
w
h
er
e
is
t
h
e
er
r
o
r
o
f
th
e
ap
p
r
o
x
i
m
ated
s
p
ee
ch
s
a
m
p
le
to
th
e
ac
t
u
al
s
p
ee
ch
s
a
m
p
le
,
is
t
h
e
co
l
u
m
n
v
ec
to
r
t
h
at
co
n
s
is
t
s
o
f
all
t
h
e
p
a
s
t
s
p
ee
c
h
s
a
m
p
les,
a
n
d
is
th
e
co
lu
m
n
v
ec
to
r
th
at
co
n
s
i
s
ts
o
f
all
t
h
e
L
P
C
co
ef
f
ici
en
ts
.
T
o
m
in
i
m
ize
,
th
en
(
2
)
is
d
if
f
er
en
tia
ted
w
it
h
r
esp
ec
t
to
an
d
eq
u
atin
g
th
e
r
esu
lt
to
ze
r
o
;
th
u
s
th
e
o
p
ti
m
u
m
L
P
C
co
ef
f
icie
n
t
s
ca
n
b
e
o
b
tain
ed
s
u
ch
t
h
at
∑
[
∑
]
,
(
3
)
w
h
er
e
is
th
e
co
v
ar
ian
ce
m
a
tr
ix
an
d
is
th
e
cr
o
s
s
-
co
r
r
elatio
n
v
ec
to
r
b
et
w
ee
n
t
h
e
p
ast
s
p
ee
ch
s
a
m
p
les
an
d
t
h
e
c
u
r
r
en
t
s
p
ee
ch
s
a
m
p
le.
Her
e,
t
h
e
L
P
C
co
ef
f
icien
ts
ar
e
u
s
ed
to
m
o
d
el
t
h
e
ch
ar
ac
ter
i
s
tic
o
f
v
o
ca
l
tr
ac
t,
w
h
er
e
th
e
s
e
co
ef
f
i
cien
ts
ar
e
f
u
r
t
h
er
to
b
e
p
r
o
ce
s
s
ed
in
t
h
e
ap
p
licatio
n
o
f
L
P
C
m
o
d
el
(
e.
g
.
s
p
ee
ch
s
y
n
t
h
esi
s
)
.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
0
8
8
-
8708
I
J
E
C
E
Vo
l.
7
,
No
.
2
,
A
p
r
il
2
0
1
7
:
84
2
–
8
49
844
3.
T
RANSF
O
RM
E
D
DO
M
AIN
L
P
C
M
O
DE
L
:
F
U
Z
Z
Y
R
E
CURS
I
V
E
L
E
AS
T
-
S
Q
UAR
E
S
AP
P
RO
ACH
Fro
m
(
1
)
,
th
e
L
P
C
m
o
d
el
i
s
a
m
o
d
el
o
f
ap
p
r
o
x
i
m
ate
cu
r
r
en
t
s
p
ee
c
h
s
a
m
p
le
w
it
h
th
e
li
n
ea
r
co
m
b
i
n
atio
n
o
f
p
ast
s
p
ee
ch
s
a
m
p
les.
I
n
s
tead
o
f
f
ee
d
i
n
g
d
ir
ec
tl
y
in
to
th
e
L
P
C
m
o
d
el,
th
ese
p
ast
s
p
ee
ch
s
a
m
p
les
ar
e
tr
an
s
f
o
r
m
ed
to
f
u
zz
y
i
n
p
u
t
s
,
w
h
er
e
ea
ch
p
ast
s
p
ee
ch
s
a
m
p
le
is
a
n
ele
m
e
n
t
in
th
e
u
n
i
v
er
s
al
o
f
d
is
co
u
r
s
e
w
h
ich
h
a
s
a
d
eg
r
ee
o
f
m
e
m
b
er
s
h
ip
in
a
p
ar
tic
u
lar
f
u
zz
y
s
et.
So
lv
i
n
g
it i
n
t
h
e
tr
a
n
s
f
o
r
m
ed
d
o
m
ai
n
o
f
f
u
zz
y
s
y
s
te
m
,
t
h
ese
f
u
zz
y
i
n
p
u
ts
f
o
r
m
a
li
n
ea
r
co
m
b
in
ati
o
n
w
it
h
f
u
zz
y
p
ar
a
m
eter
s
to
ap
p
r
o
x
im
a
te
cu
r
r
en
t
s
p
e
ec
h
s
a
m
p
le,
.
R
ed
ef
in
ed
as
th
e
tr
an
s
f
o
r
m
ed
d
o
m
ai
n
L
P
C
m
o
d
el,
cu
r
r
en
t
s
p
ee
ch
s
a
m
p
l
e
ca
n
b
e
ap
p
r
o
x
im
a
ted
s
u
c
h
th
a
t
,
(
4
)
w
h
er
e
ar
e
th
e
f
u
zz
y
p
ar
a
m
et
er
s
,
an
d
f
o
r
ar
e
th
e
f
u
zz
y
i
n
p
u
t
s
(
i.e
.
th
e
f
u
zz
if
ied
to
p
ast
s
p
ee
ch
s
am
p
les)
th
at
co
r
r
esp
o
n
d
s
to
f
u
zz
y
r
u
le.
E
m
p
lo
y
i
n
g
t
h
e
r
ed
u
ce
d
f
u
zz
y
b
asis
f
u
n
ctio
n
[
1
6
]
as th
e
f
u
zz
y
in
p
u
ts
,
t
h
en
∏
,
(
5
)
w
h
er
e
is
th
e
d
eg
r
ee
o
f
m
e
m
b
er
s
h
ip
o
f
th
at
co
r
r
esp
o
n
d
ed
to
th
e
T
r
ian
g
u
lar
-
s
h
ap
e
d
m
e
m
b
er
s
h
ip
f
u
n
ctio
n
,
at
f
u
zz
y
r
u
le.
Fo
r
n
o
tatio
n
s
i
m
p
li
cit
y
,
lets
d
en
o
te
(
5
)
as
,
th
u
s
(
4
)
ca
n
b
e
r
e
w
r
itte
n
as
,
(
6
)
w
h
er
e
cu
r
r
en
t
s
p
ee
c
h
s
a
m
p
l
e
is
ap
p
r
o
x
i
m
ated
b
y
t
h
e
lin
ea
r
co
m
b
in
at
io
n
o
f
f
u
zz
i
f
i
ed
p
ast
s
p
ee
ch
s
a
m
p
les a
n
d
f
u
zz
y
p
ar
a
m
e
ter
s
.
L
et
b
e
th
e
co
lu
m
n
v
ec
to
r
th
at
co
n
s
is
t
s
o
f
all
f
u
zz
i
f
ied
p
ast
s
p
ee
ch
s
a
m
p
les
w
it
h
r
esp
ec
t to
f
u
zz
y
r
u
le,
an
d
b
e
th
e
co
lu
m
n
v
ec
t
o
r
th
at
co
n
s
i
s
ts
o
f
all
f
u
zz
y
p
ar
a
m
eter
s
.
B
y
i
n
tr
o
d
u
cin
g
a
n
e
x
p
o
n
e
n
tial
w
e
ig
h
ti
n
g
f
o
r
g
e
tti
n
g
f
ac
to
r
,
w
h
er
e
t
h
e
i
n
f
o
r
m
atio
n
o
f
d
is
tan
t p
ast
h
as a
les
s
er
ef
f
ec
t
o
n
th
e
co
ef
f
icie
n
t
u
p
d
atin
g
,
th
e
o
b
j
ec
tiv
e
f
u
n
ctio
n
is
g
i
v
en
b
y
∑
∑
∑
,
(
7
)
w
h
er
e
h
er
e
is
th
e
er
r
o
r
o
f
th
e
ap
p
r
o
x
im
ated
s
p
ee
ch
s
a
m
p
l
e
in
th
e
tr
an
s
f
o
r
m
ed
d
o
m
ai
n
o
f
f
u
zz
y
s
y
s
te
m
to
th
e
ac
tu
al
s
p
ee
ch
s
a
m
p
le
.
Fo
llo
w
i
n
g
th
e
r
ec
u
r
s
i
v
e
least
-
s
q
u
ar
es
(
R
L
S)
ap
p
r
o
a
ch
to
s
o
l
v
e
f
o
r
th
e
f
u
zz
y
p
ar
a
m
eter
s
,
th
en
t
h
e
o
p
tim
u
m
f
u
zz
y
p
ar
a
m
eter
s
ca
n
b
e
o
b
tain
ed
in
a
s
eq
u
en
ti
al
r
ec
u
r
s
iv
e
f
o
r
m
at
s
u
c
h
th
a
t
,
(
8
)
w
h
er
e
is
t
h
e
g
ai
n
v
ec
to
r
in
w
h
ic
h
is
t
h
e
co
r
r
elatio
n
m
a
tr
ix
o
f
f
u
zz
y
in
p
u
t
s
,
an
d
is
th
e
a
p
r
io
r
i
er
r
o
r
.
Her
e,
th
e
L
P
C
m
o
d
el
is
tr
an
s
f
o
r
m
ed
in
to
f
u
zz
y
d
o
m
a
in
,
a
n
d
th
e
L
P
C
co
ef
f
icie
n
ts
ar
e
r
ep
lace
d
b
y
th
e
f
u
zz
y
p
ar
a
m
eter
s
.
I
n
s
tead
o
f
th
e
L
P
C
co
ef
f
icie
n
t
s
,
th
e
s
e
f
u
zz
y
p
ar
am
eter
s
ca
n
b
e
u
s
ed
f
o
r
f
u
r
t
h
er
p
r
o
ce
s
s
in
g
in
th
e
ap
p
licatio
n
o
f
L
P
C
m
o
d
el.
T
h
e
f
o
r
m
u
latio
n
s
f
r
o
m
(
4
)
to
(
8
)
d
ef
i
n
e
t
h
is
tr
an
s
f
o
r
m
ed
d
o
m
a
in
L
P
C
m
o
d
el
w
it
h
i
n
te
g
r
ated
f
u
zz
y
s
y
s
te
m
,
a
n
d
it
is
k
n
o
w
n
a
s
th
e
L
P
C
m
o
d
el
w
i
th
f
u
zz
y
r
ec
u
r
s
i
v
e
leas
t
-
s
q
u
ar
e
s
ap
p
r
o
ac
h
(
FR
L
S
-
L
P
C
m
o
d
el
)
.
4.
SI
M
UL
AT
I
O
N
In
th
is
s
ec
tio
n
,
th
e
L
P
C
m
o
d
el
an
d
th
e
FR
L
S
-
L
P
C
m
o
d
el
w
il
l
b
e
ev
alu
ated
in
ter
m
s
o
f
p
r
ed
ictio
n
er
r
o
r
an
d
th
e
q
u
alit
y
o
f
s
y
n
t
h
etic
s
p
ee
ch
u
s
in
g
ex
tr
ac
ted
co
ef
f
icien
ts
.
B
ased
o
n
th
e
co
n
f
i
g
u
r
atio
n
o
f
s
y
s
te
m
id
en
ti
f
icatio
n
,
th
e
co
ef
f
icie
n
t
s
o
f
t
h
e
m
o
d
el
ca
n
b
e
i
d
en
ti
f
i
ed
u
s
i
n
g
t
h
e
R
L
S
ap
p
r
o
ac
h
.
As
t
h
e
n
a
m
e
l
in
ea
r
p
r
ed
ictiv
e
s
u
g
g
est
s
,
t
h
e
e
x
tr
a
cted
co
e
f
f
ic
ien
t
s
o
f
t
h
e
m
o
d
el
f
o
r
m
a
li
n
ea
r
co
m
b
i
n
atio
n
w
ith
p
ast
s
p
ee
ch
s
a
m
p
les to
ap
p
r
o
x
i
m
a
te
th
e
c
u
r
r
en
t sp
ee
ch
s
a
m
p
le
at
a
g
i
v
e
n
ti
m
e
i
n
s
ta
n
t
.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
J
E
C
E
I
SS
N:
2
0
8
8
-
8708
F
u
z
z
y
R
ec
u
r
s
ive
Lea
s
t
-
S
q
u
a
r
es A
p
p
r
o
a
ch
in
S
p
ee
c
h
S
yst
em
I
d
en
tifi
ca
tio
n
.
.
.
.
(
K
a
h
W
a
i Ch
ea
h
)
845
I
n
t
h
is
s
i
m
u
latio
n
,
a
s
p
ee
c
h
with
t
h
e
p
r
o
n
u
n
ciatio
n
o
f
a
w
o
r
d
:
Ma
la
y
s
ia
i
s
u
s
ed
.
T
h
e
p
r
o
n
u
n
cia
tio
n
o
f
Ma
la
y
s
ia
is
/
ma
'
-
lei
-
zi
-
a/
an
d
it
co
n
tain
s
6
0
0
0
s
p
ee
ch
s
a
m
p
les
w
ith
t
h
e
s
a
m
p
li
n
g
r
ate
o
f
.
Fig
u
r
e
1
s
h
o
w
s
t
h
e
w
a
v
ef
o
r
m
an
d
s
p
ec
t
r
o
g
r
a
m
o
f
Ma
la
y
s
ia.
Fig
u
r
e
1
.
T
h
e
w
a
v
e
f
o
r
m
an
d
s
p
ec
tr
o
g
r
am
o
f
Ma
la
y
s
ia
4
.
1
.
M
e
a
s
ure
m
e
nt
Crit
er
ia
T
h
er
e
ar
e
t
w
o
m
ea
s
u
r
e
m
en
t
c
r
iter
ia
to
ev
alu
a
te
th
e
p
er
f
o
r
m
an
ce
o
f
t
h
e
m
o
d
el:
(
i)
th
e
s
q
u
ar
ed
er
r
o
r
in
Dec
ib
el
s
ca
le,
an
d
(
ii)
th
e
s
p
ec
tr
o
g
r
a
m
o
f
th
e
s
y
n
t
h
etic
s
p
ee
ch
.
Usi
n
g
t
h
e
ex
tr
ac
ted
co
ef
f
icie
n
t
s
an
d
er
r
o
r
s
p
r
o
d
u
ce
b
y
(
8
)
,
th
e
s
y
n
th
etic
s
p
ee
ch
is
co
n
s
tr
u
cted
a
n
d
co
m
p
ar
e
w
it
h
t
h
e
o
r
i
g
in
al
s
p
ee
ch
to
p
r
o
d
u
ce
th
e
an
d
th
e
s
p
ec
tr
o
g
r
a
m
.
T
h
e
f
ir
s
t
m
ea
s
u
r
e
m
en
t c
r
iter
ia
o
f
p
r
ed
ictio
n
er
r
o
r
is
g
iv
en
b
y
̃
,
(
9
)
w
h
er
e
is
th
e
b
ase
1
0
lo
g
ar
ith
m
s
ca
le,
an
d
̃
is
th
e
ap
p
r
o
x
i
m
ated
s
p
ee
ch
s
a
m
p
le
at
ti
m
e
in
s
tan
t
u
s
i
n
g
p
ast
s
p
ee
ch
s
a
m
p
le
s
(
i.e
.
̃
f
o
r
th
e
L
P
C
m
o
d
el
an
d
̃
f
o
r
th
e
FR
L
S
-
L
P
C
m
o
d
el)
.
T
h
is
is
to
co
m
p
ar
e
th
e
w
a
v
e
f
o
r
m
d
if
f
er
en
ce
b
e
t
w
ee
n
th
e
o
r
ig
i
n
al
s
p
ee
c
h
an
d
th
e
s
y
n
th
et
ic
s
p
ee
ch
.
B
esid
es
(
9
)
,
th
e
s
p
ec
t
r
o
g
r
a
m
o
f
s
y
n
t
h
etic
s
p
ee
c
h
w
ill
b
e
co
m
p
ar
ed
w
it
h
th
e
s
p
ec
tr
o
g
r
a
m
o
f
o
r
ig
i
n
al
s
p
ee
ch
i
n
o
r
d
er
to
ill
u
s
tr
ate
t
h
e
e
f
f
icien
c
y
o
f
t
h
e
e
x
tr
ac
ted
co
ef
f
icien
ts
i
n
r
ep
r
ese
n
ti
n
g
t
h
e
e
n
er
g
y
f
ill
o
f
t
h
e
s
p
ee
ch
s
a
m
p
les.
Fi
g
u
r
e
2
s
h
o
w
s
th
e
p
ar
titi
o
n
ed
s
p
ec
tr
o
g
r
a
m
o
f
Ma
la
y
s
ia.
Fig
u
r
e
2
.
T
h
e
p
ar
t
itio
n
ed
s
p
ec
tr
o
g
r
a
m
o
f
Ma
la
y
s
ia
Fro
m
Fi
g
u
r
e
2
,
th
e
co
lo
r
s
o
f
s
p
ec
tr
o
g
r
am
r
ep
r
ese
n
t
d
if
f
er
en
t
lev
els
o
f
en
er
g
y
co
n
tain
ed
i
n
a
s
p
ee
ch
.
T
h
e
r
ed
in
d
icate
s
th
e
h
i
g
h
e
s
t
en
er
g
y
w
h
er
e
t
h
e
s
o
u
n
d
o
f
a
s
p
ee
ch
is
m
ad
e,
w
h
ile
t
h
e
d
ar
k
b
lu
e
in
d
icate
s
th
e
s
ilen
t
r
eg
io
n
o
f
a
s
p
ee
ch
.
T
h
e
tr
an
s
itio
n
o
f
co
lo
r
s
f
r
o
m
h
ig
h
en
er
g
y
to
s
ilen
ce
i
s
f
r
o
m
r
ed
,
f
o
llo
w
ed
b
y
y
ello
w
an
d
lig
h
t
b
l
u
e,
f
in
a
ll
y
to
d
ar
k
b
lu
e.
T
h
e
p
r
o
n
u
n
ciatio
n
o
f
Ma
la
y
s
ia
co
n
s
is
ts
o
f
f
o
u
r
p
h
ases
,
w
h
ic
h
ar
e
in
d
icate
d
b
y
p
ar
titi
o
n
a
(
/ma
'/
)
,
p
ar
titi
o
n
b
(
/lei/
)
,
p
ar
titi
o
n
c
(
/z
i/
)
,
a
n
d
p
ar
titi
o
n
d
(
/a
/
)
,
w
it
h
th
e
en
d
i
n
g
o
f
p
ar
titi
o
n
e
(
s
ilen
ce
)
in
Fig
u
r
e
2
.
Fo
r
p
a
r
titi
o
n
a,
th
e
en
er
g
y
f
ill
is
co
n
ce
n
tr
ated
at
lo
w
f
r
eq
u
en
c
y
d
u
e
to
th
e
n
asal
co
n
s
o
n
a
n
t,
/m/
in
w
h
ic
h
it
h
ig
h
l
y
d
a
m
p
s
th
e
f
o
llo
w
i
n
g
v
o
w
el,
/a
/
.
Fo
r
p
ar
titi
o
n
b
,
th
e
p
r
o
n
u
n
ciatio
n
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
0
8
8
-
8708
I
J
E
C
E
Vo
l.
7
,
No
.
2
,
A
p
r
il
2
0
1
7
:
84
2
–
8
49
846
co
m
p
r
is
e
s
o
f
s
e
m
iv
o
w
el,
/l/
a
n
d
t
w
o
f
r
o
n
t
v
o
w
el
s
,
/e
/
a
n
d
/i/
.
D
u
e
to
th
e
tr
an
s
itio
n
al
n
a
tu
r
e
o
f
/l/
,
it
o
f
te
n
in
f
lu
e
n
ce
s
b
y
t
h
e
v
o
w
el
t
h
at
f
o
llo
w
s
,
t
h
u
s
t
h
e
e
n
er
g
y
f
il
l
i
s
co
n
ce
n
tr
ated
at
h
ig
h
a
n
d
lo
w
f
r
eq
u
en
c
y
a
s
t
h
e
n
atu
r
e
o
f
b
o
th
f
r
o
n
t v
o
w
el
s
.
F
o
r
p
ar
titi
o
n
c,
th
e
e
n
er
g
y
f
il
l is
co
n
ce
n
tr
ated
at
h
i
g
h
f
r
eq
u
en
c
y
d
u
e
to
th
e
v
o
ice
d
f
r
icati
v
e,
/z
/
an
d
t
h
e
f
r
o
n
t
v
o
wel,
/i/
i
n
w
h
ich
p
r
o
d
u
ce
ex
citat
io
n
an
d
h
i
g
h
f
r
eq
u
e
n
c
y
r
eso
n
a
n
ce
s
.
Fo
r
p
ar
titi
o
n
d
,
it
is
o
b
v
io
u
s
th
a
t
t
h
e
en
er
g
y
f
ill
i
s
co
n
ce
n
tr
ated
at
m
id
d
le
an
d
lo
w
f
r
eq
u
e
n
c
y
d
u
e
to
t
h
e
m
id
v
o
w
el,
/a
/
.
Fin
a
ll
y
,
en
er
g
y
f
ill
h
ar
d
l
y
ca
n
b
e
s
ee
n
at
p
ar
titi
o
n
e
as
it
is
th
e
en
d
in
g
o
f
t
h
e
s
p
ee
ch
.
Fo
r
f
u
r
th
er
s
p
ec
tr
o
g
r
a
m
an
al
y
s
is
,
r
ea
d
er
s
ar
e
ad
v
is
ed
a
n
d
r
ef
er
to
[
2
]
f
o
r
m
o
r
e
d
etailed
illu
s
tr
atio
n
s
a
n
d
ex
p
lan
atio
n
s
.
4
.
2
.
L
i
m
it
ed
P
a
s
t
S
peec
h Sa
m
p
le
s
I
n
t
h
is
s
i
m
u
latio
n
,
t
h
e
L
P
C
m
o
d
el
is
te
s
ted
w
it
h
1
0
p
ast
s
p
ee
ch
s
a
m
p
le
s
(
w
h
ic
h
i
s
s
u
g
g
est
ed
in
[
2
]
)
,
an
d
also
w
i
th
th
e
li
m
itatio
n
o
f
o
n
l
y
2
p
as
t
s
p
ee
c
h
s
a
m
p
les
(
th
e
m
in
i
m
u
m
n
u
m
b
er
o
f
p
ast
s
p
ee
ch
s
a
m
p
le
s
i
n
o
r
d
er
f
o
r
th
e
s
y
n
t
h
etic
s
p
ee
ch
to
b
e
in
telli
g
ib
le
[
4
]
)
.
W
h
ile
f
o
r
th
e
FR
L
S
-
L
P
C
m
o
d
el,
o
n
l
y
2
p
ast
s
p
ee
c
h
s
a
m
p
les
ar
e
u
s
ed
a
n
d
ea
ch
o
f
t
h
e
m
i
s
f
u
zz
i
f
ied
w
i
th
t
h
r
ee
T
r
ian
g
u
lar
-
s
h
ap
ed
m
e
m
b
er
s
h
ip
f
u
n
ctio
n
s
.
T
h
e
p
ar
am
eter
s
f
o
r
T
r
ian
g
u
lar
-
s
h
a
p
ed
m
e
m
b
er
s
h
ip
f
u
n
ct
io
n
ar
e
ch
o
s
e
n
ac
co
r
d
in
g
l
y
b
ased
o
n
t
h
e
r
a
n
g
e
o
f
p
ast
s
p
ee
ch
s
a
m
p
le
s
t
o
o
b
tain
r
ed
u
ce
d
f
u
zz
y
b
a
s
is
f
u
n
ctio
n
[
1
6
]
.
A
t
ev
er
y
iter
atio
n
(
i.e
.
at
ea
ch
ti
m
e
i
n
s
ta
n
t
)
,
o
n
l
y
a
li
m
ited
n
u
m
b
er
o
f
p
ast
s
p
ee
ch
s
a
m
p
les
w
ill
b
e
f
ed
i
n
to
th
e
m
o
d
el
(
i.e
.
o
n
l
y
u
n
til
ar
e
u
s
ed
to
ap
p
r
o
x
i
m
ate
)
.
Fo
r
s
u
c
h
,
b
o
th
L
P
C
m
o
d
els
ar
e
with
an
d
,
w
h
ile
f
o
r
th
e
FR
L
S
-
L
P
C
m
o
d
el
is
w
i
th
.
Fig
u
r
e
3
s
h
o
w
s
t
h
e
w
a
v
e
f
o
r
m
,
lear
n
in
g
c
u
r
v
e
o
f
er
r
o
r
an
d
s
p
ec
tr
o
g
r
a
m
o
f
th
e
s
y
n
t
h
etic
s
p
ee
ch
.
Fig
u
r
e
3
.
T
h
e
w
a
v
e
f
o
r
m
,
lear
n
in
g
c
u
r
v
e
o
f
er
r
o
r
an
d
s
p
ec
tr
o
g
r
a
m
o
f
s
y
n
t
h
etic
s
p
ee
c
h
u
s
i
n
g
th
e
L
P
C
m
o
d
el
w
it
h
(
f
ir
s
t r
o
w
)
,
t
h
e
L
P
C
m
o
d
el
w
it
h
(
s
ec
o
n
d
r
o
w
)
,
a
n
d
th
e
FR
L
S
-
L
P
C
m
o
d
el
w
it
h
(
th
ir
d
r
o
w
)
Fro
m
Fi
g
u
r
e
3
,
ev
e
n
w
i
th
th
e
li
m
i
tatio
n
o
f
2
p
as
t
s
p
e
ec
h
s
a
m
p
les,
th
e
F
R
L
S
-
L
P
C
m
o
d
el
o
u
tp
er
f
o
r
m
s
b
o
th
L
P
C
m
o
d
els
.
Fo
r
th
e
L
P
C
m
o
d
els,
b
o
th
a
r
e
at
ar
o
u
n
d
-
30
d
B
;
w
h
ile
f
o
r
th
e
F
R
L
S
-
L
P
C
m
o
d
el
,
alt
h
o
u
g
h
w
it
h
o
n
l
y
2
p
ast
s
p
ee
ch
s
a
m
p
le
s
b
ein
g
f
ee
d
in
to
th
e
m
o
d
el
,
th
e
is
at
ar
o
u
n
d
-
3
5
0dB
an
d
f
ar
lo
w
er
th
a
n
b
o
th
L
P
C
m
o
d
els.
I
n
ter
m
s
o
f
t
h
e
w
a
v
e
f
o
r
m
,
th
e
FR
L
S
-
L
P
C
m
o
d
el
clo
s
el
y
r
ese
m
b
les
th
e
o
r
ig
in
al
s
p
ee
c
h
w
av
e
f
o
r
m
a
n
d
its
s
h
ap
e
is
m
o
r
e
s
i
m
ilar
t
h
an
th
o
s
e
w
a
v
ef
o
r
m
s
r
ese
m
b
led
b
y
th
e
L
P
C
m
o
d
el
s
.
I
n
ter
m
s
o
f
t
h
e
s
p
ec
tr
o
g
r
a
m
a
n
al
y
s
i
s
,
it
ca
n
b
e
clea
r
l
y
s
ee
n
th
at
th
e
L
P
C
m
o
d
el
w
it
h
10
p
ast
s
p
ee
c
h
s
a
m
p
les
o
u
tp
er
f
o
r
m
s
t
h
e
L
P
C
m
o
d
el
w
it
h
o
n
l
y
2
p
as
t
s
p
ee
c
h
s
a
m
p
les.
T
h
e
s
y
n
t
h
etic
s
p
ee
c
h
e
n
er
g
y
f
il
l
o
f
t
h
e
L
P
C
m
o
d
el
w
it
h
o
n
l
y
2
p
ast
s
p
ee
ch
s
a
m
p
le
s
ar
e
s
p
r
ea
d
ac
r
o
s
s
an
d
n
o
t
co
n
ce
n
tr
ated
in
t
h
e
r
eg
io
n
s
w
h
er
e
it
s
h
o
u
ld
b
e
as
in
t
h
e
o
r
ig
in
a
l
s
p
ee
ch
.
I
n
th
e
ca
s
e
o
f
10
p
ast
s
p
ee
ch
s
a
m
p
les,
t
h
e
s
y
n
t
h
etic
s
p
ee
ch
en
er
g
y
f
ill
o
f
th
e
L
P
C
m
o
d
el
is
b
etter
a
n
d
it
is
clo
s
e
l
y
r
e
s
e
m
b
le
s
t
h
o
s
e
w
it
h
o
r
i
g
in
a
l
s
p
ee
ch
.
Yet,
i
f
co
m
p
ar
ed
w
it
h
t
h
e
FR
L
S
-
L
P
C
m
o
d
el,
t
h
e
F
R
L
S
-
L
P
C
m
o
d
el
i
s
b
etter
.
A
lt
h
o
u
g
h
w
it
h
o
n
l
y
2
p
ast
s
p
ee
ch
s
a
m
p
les,
t
h
e
s
y
n
th
e
tic
s
p
ee
ch
o
f
FR
L
S
-
L
P
C
m
o
d
el
r
ese
m
b
les
th
e
d
etai
ls
o
f
en
er
g
y
f
ill
a
s
i
n
t
h
e
o
r
ig
i
n
al
s
p
ee
ch
.
I
t
ca
n
b
e
s
ee
n
at
p
ar
titi
o
n
b
an
d
p
ar
titi
o
n
c,
th
e
en
er
g
y
f
illed
at
m
id
d
le
f
r
eq
u
e
n
c
y
i
s
li
g
h
t b
lu
e
f
o
r
th
e
F
R
L
S
-
L
P
C
m
o
d
el
(
w
h
ic
h
Evaluation Warning : The document was created with Spire.PDF for Python.
I
J
E
C
E
I
SS
N:
2
0
8
8
-
8708
F
u
z
z
y
R
ec
u
r
s
ive
Lea
s
t
-
S
q
u
a
r
es A
p
p
r
o
a
ch
in
S
p
ee
c
h
S
yst
em
I
d
en
tifi
ca
tio
n
.
.
.
.
(
K
a
h
W
a
i Ch
ea
h
)
847
is
s
i
m
i
lar
to
o
r
ig
i
n
al
s
p
ee
c
h
)
,
w
h
ile
it
i
s
y
e
llo
w
f
o
r
th
e
L
P
C
m
o
d
el
w
it
h
10
p
ast
s
p
ee
ch
s
a
m
p
les.
B
esid
es,
a
t
u
p
p
er
r
ig
h
t
co
r
n
er
o
f
p
ar
titi
o
n
e,
th
e
en
er
g
y
f
il
led
is
d
ar
k
b
lu
e
f
o
r
th
e
F
R
L
S
-
L
P
C
m
o
d
el
(
w
h
ic
h
is
s
i
m
ilar
to
o
r
ig
in
al
s
p
ee
ch
)
,
w
h
ile
it
i
s
l
ig
h
t
b
lu
e
f
o
r
th
e
L
P
C
m
o
d
el
w
it
h
10
p
ast
s
p
ee
ch
s
a
m
p
le
s
.
T
h
e
s
p
ec
tr
o
g
r
a
m
an
al
y
s
is
s
h
o
w
s
t
h
at
t
h
e
d
is
tr
i
b
u
tio
n
o
f
s
y
n
th
e
tic
s
p
ee
ch
en
er
g
y
f
ill
o
f
t
h
e
FR
L
S
-
L
P
C
m
o
d
el
is
clea
r
er
an
d
m
o
r
e
d
etailed
th
a
n
th
e
L
P
C
m
o
d
el
w
i
th
10
p
ast
s
p
ee
ch
s
a
m
p
les
(
n
o
t
to
m
e
n
tio
n
t
h
at
t
h
e
L
P
C
m
o
d
el
w
it
h
j
u
s
t
2
p
ast s
p
ee
ch
s
a
m
p
le
s
)
.
4
.
3
.
Co
rr
up
t
ed
P
a
s
t
Sp
ee
ch
Sa
mp
les
I
n
o
r
d
er
to
p
u
s
h
th
e
test
f
u
r
t
h
e
r
,
w
h
ite
Ga
u
s
s
ian
n
o
is
e
is
ad
d
ed
to
th
e
p
ast
s
p
ee
ch
s
am
p
le
s
w
h
er
e
th
e
s
ig
n
al
-
to
-
n
o
is
e
r
atio
(
SNR
)
d
r
o
p
s
to
1
0
d
B
.
T
h
ese
co
r
r
u
p
ted
p
ast
s
p
ee
ch
s
a
m
p
le
s
w
i
ll
b
e
f
e
d
in
to
t
h
e
m
o
d
el
to
test
t
h
e
ab
ilit
y
o
f
s
u
c
h
m
o
d
el
i
n
ex
tr
ac
tin
g
t
h
e
co
ef
f
icie
n
t
s
with
t
h
e
p
r
esen
ce
o
f
m
ea
s
u
r
e
m
e
n
t
n
o
is
e.
Fo
r
s
u
c
h
lo
w
SN
R
,
t
h
e
co
r
r
u
p
ted
s
p
ee
c
h
i
s
a
n
o
i
s
y
s
o
u
n
d
r
at
h
er
t
h
a
n
a
p
r
o
p
er
p
r
o
n
u
n
ciatio
n
o
f
Ma
l
a
y
s
ia.
I
f
th
e
SN
R
is
h
ig
h
er
,
th
e
n
t
h
e
p
r
o
n
u
n
ciatio
n
o
f
Ma
la
y
s
ia
ca
n
b
e
h
ea
r
d
,
to
g
eth
er
w
it
h
s
o
m
e
b
ac
k
g
r
o
u
n
d
n
o
is
e.
Fo
r
th
e
L
P
C
m
o
d
els,
2
a
n
d
1
0
co
r
r
u
p
ted
p
ast
s
p
ee
ch
s
a
m
p
les
w
i
ll
b
e
f
ed
i
n
to
t
h
e
m
o
d
el
;
w
h
ile
f
o
r
th
e
FR
L
S
-
L
P
C
m
o
d
els,
o
n
l
y
2
co
r
r
u
p
ted
p
ast
s
p
ee
ch
s
a
m
p
l
es
w
il
l
b
e
f
ed
i
n
to
t
h
e
m
o
d
el.
Usi
n
g
th
e
ex
tr
ac
ted
co
ef
f
icie
n
t
s
,
t
h
e
s
y
n
t
h
etic
s
p
ee
ch
h
er
e
is
th
e
r
ec
o
n
s
tr
u
cted
s
p
ee
ch
f
r
o
m
n
o
is
e
to
r
ese
m
b
le
th
e
o
r
ig
in
al
s
p
ee
ch
.
Fig
u
r
e
4
s
h
o
w
s
t
h
e
w
a
v
e
f
o
r
m
,
lear
n
in
g
c
u
r
v
e
o
f
er
r
o
r
an
d
s
p
ec
tr
o
g
r
a
m
o
f
t
h
e
s
y
n
t
h
etic
s
p
ee
ch
u
s
in
g
co
r
r
u
p
ted
p
ast s
p
ee
ch
s
a
m
p
les
w
it
h
o
n
l
y
1
0
d
B
SNR
.
Fig
u
r
e
4
.
T
h
e
w
a
v
e
f
o
r
m
,
lear
n
in
g
c
u
r
v
e
o
f
er
r
o
r
an
d
s
p
ec
tr
o
g
r
a
m
o
f
s
y
n
t
h
etic
s
p
ee
c
h
u
s
i
n
g
th
e
L
P
C
m
o
d
el
w
it
h
(
f
ir
s
t r
o
w
)
,
t
h
e
L
P
C
m
o
d
el
w
it
h
(
s
ec
o
n
d
r
o
w
)
,
t
h
e
FR
L
S
-
L
P
C
m
o
d
el
w
it
h
,
3
m
e
m
b
er
s
h
ip
f
u
n
ctio
n
s
(
th
ir
d
r
o
w
)
,
an
d
th
e
F
R
L
S
-
L
P
C
m
o
d
e
l
w
it
h
,
6
m
e
m
b
er
s
h
ip
f
u
n
ctio
n
s
(
f
o
u
r
th
r
o
w
)
Fro
m
Fi
g
u
r
e
4
,
ev
e
n
w
i
th
th
e
li
m
itatio
n
o
f
s
u
c
h
co
r
r
u
p
ted
(
1
0
d
B
SNR
)
2
p
ast
s
p
ee
ch
s
a
m
p
les,
t
h
e
FR
L
S
-
L
P
C
m
o
d
els
ar
e
o
u
tp
er
f
o
r
m
ed
b
o
th
L
P
C
m
o
d
els
in
esti
m
ati
n
g
t
h
e
cu
r
r
en
t
s
p
ee
c
h
s
a
m
p
les.
Fo
r
b
o
th
L
P
C
m
o
d
els,
th
e
s
y
n
t
h
etic
s
p
ee
ch
is
f
ailed
to
r
ec
o
n
s
tr
u
ct
t
h
e
o
r
ig
in
al
s
p
ee
ch
u
s
in
g
co
r
r
u
p
ted
p
ast
s
p
ee
ch
s
a
m
p
les
o
f
o
n
l
y
1
0
d
B
SNR
.
A
lt
h
o
u
g
h
t
h
e
o
f
b
o
th
L
P
C
m
o
d
els
ar
e
at
ar
o
u
n
d
-
3
0
d
B
,
th
e
w
a
v
e
f
o
r
m
s
ar
e
to
tall
y
u
n
r
ec
o
g
n
izab
le.
Fro
m
th
e
s
p
ec
tr
o
g
r
a
m
a
n
al
y
s
is
,
th
e
s
y
n
t
h
etic
s
p
ee
c
h
o
f
b
o
t
h
L
P
C
m
o
d
els
i
s
a
n
o
i
s
y
p
r
o
n
u
n
ciat
io
n
r
at
h
er
th
a
n
a
p
r
o
p
er
s
p
ee
ch
.
A
lt
h
o
u
g
h
t
h
e
L
P
C
m
o
d
el
w
i
th
1
0
p
ast
s
p
ee
ch
s
a
m
p
les
is
ab
le
to
r
ec
o
n
s
tr
u
ct
s
o
m
e
o
f
th
e
lo
w
f
r
eq
u
en
c
y
e
n
er
g
y
f
i
ll,
y
et
th
e
o
v
er
all
en
er
g
y
f
ill
is
f
ar
d
if
f
er
e
n
t
f
r
o
m
t
h
e
o
r
ig
in
a
l
s
p
ee
ch
.
Fo
r
b
o
th
L
P
C
m
o
d
els
,
th
is
s
i
m
u
la
tio
n
i
s
also
co
n
d
u
cted
w
it
h
th
e
co
r
r
u
p
ted
p
ast
s
p
ee
ch
s
a
m
p
les
o
f
3
0
d
B
SNR
: th
e
s
y
n
t
h
etic
s
p
ee
ch
b
ec
o
m
es
b
etter
w
h
er
e
t
h
e
p
r
o
n
u
n
c
iatio
n
o
f
Ma
la
y
s
ia
ca
n
b
e
h
ea
r
d
w
i
th
s
o
m
e
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
0
8
8
-
8708
I
J
E
C
E
Vo
l.
7
,
No
.
2
,
A
p
r
il
2
0
1
7
:
84
2
–
8
49
848
b
ac
k
g
r
o
u
n
d
n
o
is
e
.
R
es
u
lt
s
s
h
o
w
th
a
t
t
h
e
p
er
f
o
r
m
an
ce
o
f
L
P
C
m
o
d
el
d
ep
en
d
s
o
n
t
h
e
a
v
ail
ab
ilit
y
o
f
clea
n
p
ast
s
p
ee
ch
s
a
m
p
les.
Fo
r
b
o
th
FR
L
S
-
L
P
C
m
o
d
els,
th
e
is
a
t
ar
o
u
n
d
-
4
0
d
B
f
o
r
th
r
ee
m
e
m
b
er
s
h
ip
f
u
n
c
tio
n
s
an
d
a
t
ar
o
u
n
d
-
6
0
d
B
f
o
r
s
ix
m
e
m
b
er
s
h
ip
f
u
n
ct
io
n
s
.
Fro
m
t
h
e
w
av
ef
o
r
m
s
o
f
s
y
n
t
h
etic
s
p
ee
c
h
u
s
in
g
F
R
L
S
-
L
P
C
m
o
d
el
s
,
th
e
s
h
ap
es
ar
e
clo
s
el
y
m
atc
h
ed
to
th
e
w
av
e
f
o
r
m
o
f
o
r
ig
i
n
al
s
p
ee
ch
,
esp
ec
iall
y
t
h
e
o
n
e
w
it
h
h
i
g
h
er
n
u
m
b
er
o
f
m
e
m
b
er
s
h
ip
f
u
n
cti
o
n
s
.
Fro
m
th
e
s
p
ec
tr
o
g
r
a
m
an
al
y
s
i
s
,
t
h
e
d
is
tr
ib
u
tio
n
o
f
s
y
n
t
h
etic
s
p
ee
ch
e
n
er
g
y
f
ill
is
clea
r
er
a
n
d
co
m
p
ar
ab
le
to
th
e
o
r
ig
in
al
s
p
ee
ch
a
lth
o
u
g
h
it
i
s
r
o
u
g
h
l
y
i
n
s
h
ap
e.
W
it
h
h
ig
h
er
n
u
m
b
er
o
f
m
e
m
b
er
s
h
ip
f
u
n
ctio
n
s
to
f
u
zz
if
y
t
h
e
co
r
r
u
p
ted
p
ast
s
p
ee
ch
s
a
m
p
les,
t
h
e
q
u
alit
y
o
f
s
y
n
th
e
tic
s
p
ee
ch
b
ec
o
m
e
s
b
etter
.
I
t
ca
n
b
e
s
ee
n
at
th
e
s
p
ec
tr
o
g
r
am
o
f
s
y
n
t
h
etic
s
p
ee
ch
u
s
in
g
F
R
L
S
-
L
P
C
m
o
d
el
w
i
th
s
i
x
m
e
m
b
er
s
h
ip
f
u
n
ctio
n
s
:
th
e
lo
w
f
r
eq
u
e
n
c
y
en
er
g
y
f
ill
at
p
ar
titi
o
n
a,
th
e
l
o
w
a
n
d
h
i
g
h
f
r
eq
u
e
n
c
y
en
er
g
y
f
ill
at
p
ar
titi
o
n
b
,
th
e
h
ig
h
f
r
eq
u
en
c
y
e
n
er
g
y
f
ill
at
p
ar
titi
o
n
c,
an
d
th
e
lo
w
a
n
d
m
id
d
le
f
r
eq
u
e
n
c
y
e
n
er
g
y
f
il
l
at
p
ar
titi
o
n
d
ar
e
m
atc
h
ed
w
it
h
t
h
e
o
r
ig
i
n
al
s
p
ee
ch
.
P
la
y
i
n
g
it
as
a
u
d
io
,
th
e
p
r
o
n
u
n
c
iatio
n
o
f
Ma
la
y
s
ia
ca
n
b
e
h
ea
r
d
clea
r
l
y
.
I
n
th
is
s
i
m
u
latio
n
,
it
i
s
o
b
v
io
u
s
t
h
at
t
h
e
FR
L
S
-
L
P
C
m
o
d
el
o
u
t
p
er
f
o
r
m
ed
t
h
e
L
P
C
m
o
d
el.
E
v
en
w
it
h
o
n
l
y
2
s
u
c
h
co
r
r
u
p
ted
p
ast s
p
ee
ch
s
a
m
p
les
o
f
1
0
d
B
SNR
,
th
e
FR
L
S
-
L
P
C
m
o
d
el
is
ab
le
to
r
ec
o
n
s
tr
u
ct
t
h
e
s
p
ee
ch
s
u
c
h
t
h
at
it c
lo
s
el
y
r
ese
m
b
les t
h
e
o
r
ig
i
n
al
s
p
ee
ch
; c
o
m
p
ar
ed
w
i
th
t
h
o
s
e
u
s
i
n
g
L
P
C
m
o
d
el
s
ar
e
f
ailed
.
5.
CO
NCLU
SI
O
N
I
n
t
h
is
p
ap
er
,
f
u
zz
y
s
y
s
te
m
i
s
d
ir
ec
tl
y
i
n
te
g
r
ated
i
n
to
t
h
e
L
P
C
m
o
d
el
u
s
i
n
g
r
ec
u
r
s
iv
e
lea
s
t
-
s
q
u
ar
es
ap
p
r
o
ac
h
to
cr
ea
te
th
e
FR
L
S
-
L
P
C
m
o
d
el.
I
n
th
is
tr
an
s
f
o
r
m
ed
d
o
m
ain
L
P
C
m
o
d
el,
f
u
zz
y
p
ar
am
eter
s
ar
e
u
s
ed
to
ap
p
r
o
x
i
m
ate
t
h
e
c
u
r
r
en
t
s
p
ee
ch
s
a
m
p
le,
i
n
w
h
ich
its
p
er
f
o
r
m
an
ce
d
ep
en
d
s
o
n
t
h
e
f
u
zz
y
r
u
le
s
a
n
d
m
e
m
b
er
s
h
ip
f
u
n
ctio
n
s
r
ath
er
t
h
an
o
n
t
h
e
n
u
m
b
er
o
f
p
ast
s
p
e
ec
h
s
a
m
p
le
s
.
Alth
o
u
g
h
t
h
e
co
m
p
u
tat
io
n
o
f
f
u
zz
y
in
p
u
t
s
r
eq
u
ir
es
ad
d
itio
n
a
l
co
m
p
u
tatio
n
al
co
s
t
co
m
p
ar
ed
to
th
e
L
P
C
m
o
d
el,
h
o
w
ev
er
t
h
e
r
es
u
lts
ar
e
s
ig
n
i
f
ica
n
t
en
o
u
g
h
to
m
a
k
e
a
tr
ad
e
o
f
f
.
Si
m
u
latio
n
s
h
o
w
s
t
h
at
alt
h
o
u
g
h
w
it
h
l
i
m
ited
n
u
m
b
er
o
f
p
ast
s
p
ee
ch
s
a
m
p
les
f
ed
,
th
e
s
y
n
t
h
etic
s
p
ee
ch
o
b
tain
ed
b
y
t
h
e
F
R
L
S
-
L
P
C
m
o
d
el
i
s
f
a
r
b
etter
th
a
n
t
h
o
s
e
o
f
th
e
L
P
C
m
o
d
el
w
h
ic
h
h
a
v
e
s
u
f
f
ici
e
n
t
n
u
m
b
er
o
f
p
as
t
s
p
ee
ch
s
a
m
p
les;
i
n
ter
m
s
o
f
p
r
ed
ictio
n
er
r
o
r
an
d
s
p
ec
tr
o
g
r
a
m
an
al
y
s
is
.
T
h
e
s
i
m
u
lat
io
n
i
s
ev
e
n
te
s
ted
w
it
h
co
r
r
u
p
ted
p
ast
s
p
ee
ch
s
a
m
p
l
es,
an
d
t
h
e
r
es
u
lt
s
h
o
w
s
th
a
t
ev
en
w
it
h
s
u
ch
lo
w
SNR
m
o
d
el
f
ed
,
t
h
e
F
R
L
S
-
L
P
C
m
o
d
el
h
as
p
r
o
v
e
n
it
s
p
er
f
o
r
m
a
n
ce
to
r
ese
m
b
le
th
e
o
r
i
g
in
a
l
s
p
ee
ch
.
B
o
th
s
i
m
u
lat
io
n
s
s
h
o
w
t
h
e
v
iab
ilit
y
o
f
F
R
L
S
-
L
P
C
m
o
d
el
in
s
u
c
h
co
n
s
tr
icted
co
n
d
itio
n
;
w
h
ile
th
e
L
P
C
m
o
d
el
i
s
u
n
d
er
p
er
f
o
r
m
ed
.
Si
n
ce
th
e
p
er
f
o
r
m
an
ce
o
f
FR
L
S
-
L
P
C
m
o
d
el
d
ep
en
d
s
o
n
t
h
e
f
u
zz
y
r
u
les
r
ath
er
t
h
a
n
t
h
e
n
u
m
b
er
o
f
p
ast
s
p
ee
ch
s
a
m
p
le
s
,
t
h
e
f
u
zz
y
p
ar
a
m
eter
s
e
x
tr
ac
t
ed
u
s
i
n
g
F
R
L
S
-
L
P
C
m
o
d
el
ca
n
b
e
a
n
al
ter
n
ati
v
e
to
th
e
L
P
C
co
ef
f
icie
n
t
s
in
t
h
e
ap
p
licatio
n
o
f
L
P
C
m
o
d
el.
ACK
NO
WL
E
D
G
E
M
E
NT
S
T
h
e
au
th
o
r
s
w
o
u
ld
li
k
e
to
ac
k
n
o
w
led
g
e
Un
iv
er
s
iti
Sai
n
s
Ma
la
y
s
ia
a
n
d
T
h
e
Min
is
tr
y
o
f
H
ig
h
er
E
d
u
ca
tio
n
Ma
la
y
s
ia
f
o
r
th
e
f
in
an
cial
s
u
p
p
o
r
t th
r
o
u
g
h
T
h
e
Fu
n
d
a
m
e
n
tal
R
esear
ch
Gr
a
n
t Sc
h
e
m
e.
RE
F
E
R
E
NC
E
S
[1
]
J.
W
.
P
ico
n
e
,
“
S
ig
n
a
l
M
o
d
e
li
n
g
T
e
c
h
n
iq
u
e
s
in
S
p
e
e
c
h
Rec
o
g
n
it
io
n
,
”
P
ro
c
e
e
d
in
g
s
o
f
th
e
IEE
E
,
v
o
l
/i
ss
u
e
:
81
(
9
)
,
p
p
.
1
2
1
5
-
1
2
4
7
,
1
9
9
3
.
[2
]
L
.
Ra
b
in
e
r
a
n
d
B
.
H.
J
u
a
n
g
,
“
F
u
n
d
a
m
e
n
tals o
f
S
p
e
e
c
h
Re
c
o
g
n
it
io
n
,”
P
re
n
ti
c
e
Ha
ll
,
1
9
9
3
.
[3
]
H.
F
.
S
il
v
e
r
m
a
n
a
n
d
N.
R.
Di
x
o
n
,
“
A
Pa
ra
me
trica
ll
y
Co
n
tro
l
led
S
p
e
c
tra
l
An
a
lys
is
S
y
ste
m
fo
r
S
p
e
e
c
h
,
”
IEE
E
T
ra
n
sa
c
ti
o
n
s o
n
A
c
o
u
stics
,
S
p
e
e
c
h
a
n
d
S
ig
n
a
l
P
ro
c
e
ss
in
g
,
v
o
l
/i
ss
u
e
:
22
(
5
)
,
p
p
.
3
6
2
-
3
8
1
,
1
9
7
4
.
[4
]
B.
S
.
A
tal
a
n
d
S
.
L
.
Ha
n
a
u
e
r,
“
S
p
e
e
c
h
An
a
lys
is
a
n
d
S
y
n
t
h
e
sis
b
y
L
i
n
e
a
r
Pre
d
icti
o
n
o
f
t
h
e
S
p
e
e
c
h
W
a
v
e
,
”
Jo
u
rn
a
l
o
f
th
e
A
c
o
u
stica
l
S
o
c
iety
o
f
Am
e
ric
a
,
v
o
l
/i
ss
u
e
:
50
(
2
)
,
p
p
.
6
3
7
-
6
5
5
,
1
9
7
1
.
[5
]
E.
S
.
G
.
Ca
ro
tt
i,
e
t
a
l.
,
“
L
in
e
a
r
Pre
d
ic
ti
v
e
Co
d
in
g
o
f
M
y
o
e
lec
tri
c
S
ig
n
a
ls
,
”
in
IE
EE
In
ter
n
a
ti
o
n
a
l
Co
n
f
e
re
n
c
e
o
n
A
c
o
u
stics
,
S
p
e
e
c
h
,
a
n
d
S
ig
n
a
l
P
r
o
c
e
ss
in
g
,
p
p
.
6
2
9
-
6
3
2
,
2
0
0
5
.
[6
]
T
.
M
.
T
h
a
sle
e
m
a
,
e
t
a
l.
,
“
M
a
la
y
a
la
m V
o
we
l
Rec
o
g
n
i
ti
o
n
Ba
se
d
o
n
L
in
e
a
r P
re
d
ictive
Co
d
i
n
g
Pa
r
a
m
e
ter
s a
n
d
K
-
NN
Al
g
o
rit
h
m
,
”
i
n
Co
n
f
e
re
n
c
e
o
n
Co
m
p
u
tatio
n
a
l
In
telli
g
e
n
t
a
n
d
M
u
lt
i
m
e
d
ia A
p
p
li
c
a
ti
o
n
s
,
p
p
.
3
6
1
-
3
6
5
,
2
0
0
7
.
[7
]
R.
S
.
H.
Iste
p
a
n
ian
,
e
t
a
l.
,
“
L
i
n
e
a
r
P
re
d
ictiv
e
Co
d
i
n
g
f
o
r
En
h
a
n
c
e
d
M
icro
a
rra
y
Da
ta
Clu
ste
rin
g
,
”
in
IEE
E
In
ter
n
a
t
io
n
a
l
W
o
rk
sh
o
p
o
n
Ge
n
o
mic
S
ig
n
a
l
Pro
c
e
ss
in
g
a
n
d
S
ta
ti
st
ics
,
p
p
.
1
-
4
,
2
0
0
7
.
[8
]
N.
T
h
e
e
ra
-
U
m
p
o
n
,
e
t
a
l.
,
“
Da
ta
r
e
c
o
n
stru
c
ti
o
n
fo
r
M
issin
g
E
lec
tro
c
a
rd
io
g
ra
m
u
si
n
g
L
i
n
e
a
r
Pre
d
icti
v
e
Co
d
in
g
,
”
i
n
IEE
E
In
tern
a
ti
o
n
a
l
Co
n
f
e
re
n
c
e
o
n
M
e
c
h
a
tro
n
ics
a
n
d
A
u
to
m
a
ti
o
n
,
p
p
.
6
3
8
-
6
4
3
,
2
0
0
8
.
[9
]
N.
F
o
ro
u
z
a
n
f
a
r,
e
t
a
l.
,
“
Dy
n
a
mic
T
e
x
tu
r
e
S
e
g
me
n
ta
ti
o
n
u
si
n
g
L
in
e
a
r
Pre
d
ictive
Co
d
in
g
,
”
in
7
th
Ira
n
ian
Co
n
f
e
re
n
c
e
o
n
M
a
c
h
in
e
V
isi
o
n
a
n
d
Im
a
g
e
P
ro
c
e
ss
in
g
,
p
p
.
1
-
5
,
2
0
1
1
.
[1
0
]
R.
J.
Ja
v
ier
a
n
d
Y.
Kim
,
“
Ap
p
li
c
a
ti
o
n
o
f
L
in
e
a
r
Pre
d
ictive
Co
d
i
n
g
fo
r
H
u
ma
n
Activity
Cla
ss
if
ic
a
ti
o
n
B
a
se
d
o
n
M
icr
o
d
o
p
p
ler
S
i
g
n
a
tu
re
s
,
”
IEE
E
G
e
o
sc
i
e
n
c
e
a
n
d
Re
m
o
te S
e
n
sin
g
L
e
tt
e
rs
,
v
o
l
/i
ss
u
e
:
11
(
10
)
,
p
p
.
1
8
3
1
-
1
8
3
4
,
2
0
1
1
.
[1
1
]
J.
D.
M
a
rk
e
l
a
n
d
A
.
H.
G
ra
y
,
“
L
in
e
a
r
P
re
d
ictio
n
o
f
S
p
e
e
c
h
,”
S
p
ri
n
g
e
r
-
V
e
rlag
,
1
9
7
6
.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
J
E
C
E
I
SS
N:
2
0
8
8
-
8708
F
u
z
z
y
R
ec
u
r
s
ive
Lea
s
t
-
S
q
u
a
r
es A
p
p
r
o
a
ch
in
S
p
ee
c
h
S
yst
em
I
d
en
tifi
ca
tio
n
.
.
.
.
(
K
a
h
W
a
i Ch
ea
h
)
849
[1
2
]
E.
Jo
n
e
s,
e
t
a
l.
,
“
Re
c
o
v
e
r
y
o
f
L
S
P
c
o
e
f
f
ici
e
n
ts
in
Vo
I
P
S
y
ste
m
s
u
sin
g
Ev
o
lv
in
g
T
a
k
a
g
i
S
u
g
e
n
o
F
u
z
z
y
M
o
d
e
ls,
”
i
n
In
ter
n
a
t
io
n
a
l
S
y
mp
o
si
u
m o
n
Ev
o
l
v
in
g
F
u
zz
y
S
y
ste
ms
,
p
p
.
2
2
1
-
2
2
6
,
2
0
0
6
.
[1
3
]
T
.
D.
P
h
a
m
,
“
Ca
n
c
e
r
Cla
ss
if
ica
ti
o
n
b
y
M
in
imizi
n
g
Fu
zz
y
S
c
a
tt
e
ri
n
g
Ef
fec
ts
,
”
in
I
EE
E
In
tern
a
ti
o
n
a
l
Co
n
f
e
re
n
c
e
o
f
F
u
z
z
y
S
y
st
e
m
s
,
p
p
.
3
7
7
-
3
8
0
,
2
0
0
8
.
[1
4
]
M
.
Jo
h
n
n
y
a
n
d
J.
M
irza
e
e
,
“
A
F
u
z
z
y
B
a
se
d
V
e
r
y
L
o
w
Bit
Ra
t
e
S
p
e
e
c
h
Co
d
i
n
g
w
it
h
Hig
h
Ac
c
u
ra
c
y
,
”
in
20
th
Ira
n
ia
n
C
o
n
fer
e
n
c
e
o
n
El
e
c
trica
l
En
g
i
n
e
e
rin
g
,
p
p
.
1
3
0
2
-
1
3
0
6
,
2
0
1
2
.
[1
5
]
L
.
X
.
W
a
n
g
a
n
d
J.
M
.
M
e
n
d
e
l,
“
F
u
z
z
y
Ba
sis
F
u
n
c
ti
o
n
,
Un
iv
e
rsa
l
A
p
p
ro
x
i
m
a
ti
o
n
,
a
n
d
Orth
o
g
o
n
a
l
L
e
a
st
-
S
q
u
a
re
s
L
e
a
rn
in
g
,
”
IEE
E
T
ra
n
s
a
c
ti
o
n
s o
n
Ne
u
ra
l
Ne
two
rk
s
,
v
o
l
/i
ss
u
e
:
3
(
5
)
,
p
p
.
8
0
7
-
8
1
4
,
1
9
9
2
.
[1
6
]
K.
W
.
Ch
e
a
h
a
n
d
N.
A
.
A
h
m
a
d
,
“
Un
iv
e
rsa
l
A
p
p
ro
x
im
a
ti
o
n
o
f
Re
d
u
c
e
d
F
u
z
z
y
Ba
sis
F
u
n
c
ti
o
n
w
it
h
Ru
sp
i
n
i
P
a
rti
t
io
n
in
g
,
”
B
u
ll
e
ti
n
o
f
th
e
M
a
l
a
y
sia
n
M
a
th
e
ma
ti
c
a
l
S
c
ien
c
e
s S
o
c
iety
,
In
P
re
ss
,
2
0
1
5
.
B
I
O
G
RAP
H
I
E
S
O
F
AUTH
O
RS
K
a
h
Wa
i
Ch
e
a
h
re
c
e
iv
e
d
a
Ba
c
h
e
lo
r
d
e
g
re
e
o
f
Bu
sin
e
ss
M
a
th
e
m
a
ti
c
s
f
ro
m
Un
iv
e
r
siti
Uta
ra
M
a
la
y
sia
in
2
0
1
2
,
a
n
d
a
M
a
st
e
r
d
e
g
re
e
o
f
S
c
ien
c
e
in
M
a
th
e
m
a
ti
c
s
f
ro
m
Un
iv
e
rsiti
S
a
in
s
M
a
la
y
sia
in
2
0
1
3
.
Cu
rre
n
tl
y
,
h
e
i
s
p
u
rsu
in
g
Do
c
to
ra
l
d
e
g
re
e
in
M
a
th
e
m
a
ti
c
s
a
t
Un
iv
e
rsiti
S
a
in
s
M
a
la
y
sia
.
His
to
p
ic
in
tere
sts
a
re
f
u
z
z
y
s
y
ste
m
a
n
d
a
d
a
p
ti
v
e
sy
ste
m
id
e
n
ti
f
ica
ti
o
n
in
w
h
ich
in
v
o
lv
e
s
c
o
m
p
u
tatio
n
a
l
a
lg
e
b
ra
in
th
e
f
ield
o
f
sig
n
a
l
p
r
o
c
e
ss
in
g
.
No
o
r
Atin
a
h
A
h
m
a
d
r
e
c
e
i
v
e
d
a
Ba
c
h
e
lo
r
d
e
g
re
e
in
M
a
th
e
m
a
ti
c
s
f
ro
m
th
e
Un
iv
e
rsit
y
o
f
Bristo
l
in
1
9
9
3
a
n
d
a
Do
c
to
ra
l
d
e
g
re
e
f
ro
m
th
e
Un
iv
e
rsity
o
f
S
o
u
th
a
m
p
to
n
i
n
1
9
9
7
.
S
h
e
jo
in
e
d
th
e
S
c
h
o
o
l
o
f
M
a
th
e
m
a
ti
c
a
l
S
c
ien
c
e
s,
U
n
iv
e
rsiti
S
a
in
s
M
a
lay
sia
in
1
9
9
7
a
n
d
b
e
c
a
m
e
a
n
a
ss
o
c
iate
p
ro
f
e
ss
o
r
in
2
0
1
4
.
He
r
c
u
rre
n
t
re
se
a
rc
h
in
tere
sts
a
re
in
a
d
a
p
ti
v
e
a
lg
o
rit
h
m
s
a
n
d
m
a
tri
x
m
e
th
o
d
s
in
sig
n
a
l
p
ro
c
e
ss
in
g
.
Evaluation Warning : The document was created with Spire.PDF for Python.