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.
11
,
No
.
2
,
A
p
r
il
2
0
2
1
,
p
p
.
1
2
5
9
~1
2
6
4
I
SS
N:
2
0
8
8
-
8708
,
DOI
: 1
0
.
1
1
5
9
1
/
i
j
ec
e
.
v
1
1
i
2
.
p
p
1
2
5
9
-
1
2
6
4
1259
J
o
ur
na
l ho
m
ep
a
g
e
:
h
ttp
:
//ij
ec
e.
ia
esco
r
e.
co
m
Speech
e
m
o
tion r
ecog
nition ba
sed
o
n SVM a
nd
K
N
N
cla
ss
ificatio
ns fus
io
n
M
o
ha
m
m
e
d J
a
w
a
d Al
Duj
a
i
li
1
,
Abba
s
E
bra
hi
m
i
-
M
o
g
ha
da
m
2
,
Ah
m
ed
F
a
t
la
w
i
3
1
De
p
a
rt
m
e
n
t
o
f
El
e
c
tro
n
ic an
d
Co
m
m
u
n
ica
ti
o
n
,
F
a
c
u
lt
y
o
f
En
g
in
e
e
rin
g
,
Un
iv
e
rsity
o
f
Ku
f
a
,
N
a
ja
f
,
Ira
q
2
El
e
c
tri
c
a
l
E
n
g
in
e
e
rin
g
De
p
a
rtm
e
n
t
F
a
c
u
lt
y
o
f
En
g
in
e
e
rin
g
,
F
e
rd
o
w
si Un
iv
e
rsit
y
o
f
M
a
sh
h
a
d
,
M
a
sh
h
a
d
,
Ira
n
3
Co
m
p
u
ter E
n
g
in
e
e
rin
g
De
p
a
rtm
e
n
t
F
a
c
u
lt
y
o
f
En
g
in
e
e
rin
g
,
F
e
rd
o
w
si Un
iv
e
rsit
y
o
f
M
a
sh
h
a
d
,
M
a
sh
h
a
d
,
Ira
n
Art
icle
I
nfo
AB
ST
RAC
T
A
r
ticle
his
to
r
y:
R
ec
eiv
ed
Oct
2
,
2
0
1
9
R
ev
i
s
ed
A
u
g
2
3
,
2
0
2
0
A
cc
ep
ted
Oct
2
1
,
2
0
2
0
Re
c
o
g
n
izin
g
th
e
se
n
se
o
f
sp
e
e
c
h
is
o
n
e
o
f
th
e
m
o
st
a
c
ti
v
e
re
se
a
r
c
h
t
o
p
ics
i
n
sp
e
e
c
h
p
ro
c
e
ss
in
g
a
n
d
in
h
u
m
a
n
-
c
o
m
p
u
ter
in
tera
c
ti
o
n
p
ro
g
ra
m
s.
De
sp
it
e
a
w
id
e
ra
n
g
e
o
f
stu
d
ies
in
th
is
sc
o
p
e
,
th
e
re
is
stil
l
a
lo
n
g
g
a
p
a
m
o
n
g
th
e
n
a
tu
ra
l
f
e
e
li
n
g
s
o
f
h
u
m
a
n
s
a
n
d
th
e
p
e
rc
e
p
ti
o
n
o
f
th
e
c
o
m
p
u
ter.
In
g
e
n
e
ra
l,
a
se
n
so
ry
re
c
o
g
n
it
io
n
sy
ste
m
f
ro
m
sp
e
e
c
h
c
a
n
b
e
d
iv
id
e
d
in
to
t
h
re
e
m
a
i
n
se
c
ti
o
n
s:
a
tt
rib
u
te
e
x
trac
ti
o
n
,
f
e
a
tu
re
se
l
e
c
ti
o
n
,
a
n
d
c
las
sif
ica
ti
o
n
.
In
t
h
is
p
a
p
e
r
,
f
e
a
tu
re
s
o
f
f
u
n
d
a
m
e
n
tal
f
r
e
q
u
e
n
c
y
(F
EZ
)
(F
0
),
e
n
e
r
g
y
(E)
,
z
e
ro
-
c
r
o
ss
in
g
ra
te
(ZCR),
f
o
u
rier
p
a
ra
m
e
ter
(F
P
),
a
n
d
v
a
rio
u
s
c
o
m
b
in
a
ti
o
n
s
o
f
th
e
m
a
r
e
e
x
trac
ted
f
ro
m
th
e
d
a
ta
v
e
c
to
r,
T
h
e
n
,
t
h
e
p
rin
c
i
p
a
l
c
o
m
p
o
n
e
n
t
a
n
a
ly
sis
(P
CA
)
a
lg
o
rit
h
m
is
u
se
d
to
re
d
u
c
e
th
e
n
u
m
b
e
r
o
f
fe
a
tu
re
s.
T
o
e
v
a
lu
a
te
th
e
s
y
ste
m
p
e
r
f
o
rm
a
n
c
e
.
T
h
e
f
u
sio
n
o
f
e
a
c
h
e
m
o
ti
o
n
a
l
sta
te
w
il
l
b
e
p
e
rf
o
r
m
e
d
late
r
u
sin
g
su
p
p
o
r
t
v
e
c
to
r
m
a
c
h
i
n
e
(S
V
M
),
K
-
n
e
a
re
st
n
e
ig
h
b
o
r
(
KN
N),
In
term
s
o
f
c
o
m
p
a
riso
n
,
si
m
il
a
r
e
x
p
e
rime
n
ts
h
a
v
e
b
e
e
n
p
e
r
f
o
rm
e
d
o
n
t
h
e
e
m
o
ti
o
n
a
l
sp
e
e
c
h
o
f
th
e
G
e
r
m
a
n
lan
g
u
a
g
e
,
En
g
li
sh
lan
g
u
a
g
e
,
a
n
d
sig
n
if
ica
n
t
re
su
lt
s w
e
re
o
b
tain
e
d
b
y
th
e
se
c
o
m
p
a
riso
n
s.
K
ey
w
o
r
d
s
:
FEZ
FP
KNN
P
C
A
Sp
ee
ch
e
m
o
tio
n
SVM
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
BY
-
SA
li
c
e
n
se
.
C
o
r
r
e
s
p
o
nd
ing
A
uth
o
r
:
Mo
h
a
m
m
ed
J
a
w
ad
A
l D
u
j
aili
Dep
ar
t
m
en
t o
f
E
lectr
o
n
ic
an
d
C
o
m
m
u
n
ica
tio
n
E
n
g
i
n
ee
r
in
g
Facu
lt
y
o
f
E
n
g
i
n
ee
r
i
n
g
,
U
n
i
v
e
r
s
it
y
o
f
K
u
f
a
Naj
af
,
I
r
aq
E
m
ail:
Mo
h
a
m
m
ed
.
ch
al
lab
@
u
o
k
u
f
a.
ed
u
.
iq
1.
I
NT
RO
D
UCT
I
O
N
Dete
ctio
n
o
f
s
e
n
s
a
tio
n
f
r
o
m
s
p
ee
ch
s
ig
n
al
i
s
o
n
e
o
f
t
h
e
r
elativ
el
y
n
e
w
b
r
an
c
h
e
s
in
s
p
ee
ch
p
r
o
ce
s
s
in
g
,
w
h
ich
ca
n
p
la
y
a
n
i
m
p
o
r
tan
t
r
o
le
in
h
u
m
a
n
-
r
o
b
o
t
in
ter
ac
tio
n
s
.
Dete
ct
in
g
th
e
s
e
n
s
e
o
f
s
p
ee
ch
is
v
er
y
u
s
e
f
u
l
f
o
r
ap
p
licatio
n
s
t
h
at
R
eq
u
ir
es
a
n
at
u
r
al
in
ter
ac
tio
n
b
et
w
ee
n
h
u
m
a
n
s
an
d
t
h
e
m
ac
h
in
e,
s
u
c
h
as
m
o
v
ie
s
,
w
eb
an
d
co
m
p
u
ter
tr
ain
i
n
g
ap
p
licatio
n
s
,
w
h
er
e
th
e
r
esp
o
n
s
e
d
ep
en
d
s
o
n
t
h
e
u
s
er
'
s
f
ee
li
n
g
s
.
I
t
ca
n
also
b
e
u
s
ed
as
a
u
s
ef
u
l
m
ed
ical
d
ev
ice
f
o
r
t
h
e
d
iag
n
o
s
is
o
f
ce
r
tai
n
p
atie
n
ts
,
s
u
ch
as
a
u
tis
m
,
P
ar
k
in
s
o
n
'
s
d
is
ea
s
e
[
1
]
.
Sp
ee
ch
r
e
co
g
n
iti
o
n
s
y
s
te
m
s
d
etec
t
th
e
e
m
o
ti
o
n
al
s
tate
o
f
t
h
e
s
p
ea
k
er
to
an
al
y
ze
th
e
s
o
u
n
d
ch
ar
ac
ter
is
tic
s
.
Ho
w
ev
er
,
h
u
m
an
v
o
ice
h
as
a
co
m
b
i
n
atio
n
o
f
i
n
f
o
r
m
atio
n
i
n
cl
u
d
in
g
n
ar
r
ativ
e
tr
ait
s
a
n
d
lex
ical,
c
u
ltu
r
al,
p
s
y
c
h
o
lo
g
ic
al,
an
d
e
m
o
tio
n
al
f
ea
t
u
r
es.
T
h
e
p
r
esen
ce
o
f
t
h
e
s
e
co
m
m
u
n
icatio
n
d
i
m
e
n
s
io
n
s
ca
u
s
e
t
h
e
v
ar
iab
le
s
th
at
a
f
f
ec
t
th
e
p
er
f
o
r
m
a
n
ce
o
f
t
h
e
d
iag
n
o
s
tic
s
y
s
te
m
.
He
n
ce
,
t
h
e
cr
ea
tio
n
o
f
e
m
o
tio
n
a
l
m
o
d
el
s
r
eq
u
ir
es
t
h
eir
ca
r
ef
u
l
co
n
s
id
er
atio
n
to
co
m
p
e
n
s
ate
t
h
e
ef
f
ec
t
s
o
f
t
h
ese
v
ar
iab
les
[
2
]
.
Hu
m
a
n
s
p
ee
c
h
co
n
s
is
ts
o
f
t
w
o
p
ar
ts
o
f
t
h
e
co
n
ten
t
an
d
th
e
to
n
e.
Hu
m
a
n
b
ein
g
s
u
s
u
a
ll
y
m
ea
n
ea
ch
o
th
er
b
o
th
b
y
co
n
te
n
t
an
d
b
y
u
s
i
n
g
t
h
e
to
n
e.
So
,
t
h
e
s
a
m
e
co
n
ten
t
t
h
at
is
ex
p
r
ess
ed
in
t
w
o
d
i
f
f
er
en
t
ten
s
es
m
a
y
h
a
v
e
t
w
o
d
if
f
er
en
t
m
ea
n
in
g
s
a
n
d
m
ea
n
i
n
g
s
.
I
n
v
e
s
tig
a
tio
n
s
o
n
th
e
r
ec
o
g
n
it
io
n
o
f
e
m
o
tio
n
f
r
o
m
s
p
ee
ch
ca
n
b
e
s
tu
d
ied
in
ter
m
s
o
f
th
e
f
ea
t
u
r
es
u
s
ed
a
n
d
t
h
e
cla
s
s
if
icatio
n
al
g
o
r
ith
m
.
Ma
n
y
r
ese
ar
ch
er
s
i
n
t
h
i
s
ar
ea
h
av
e
f
o
c
u
s
ed
th
eir
e
f
f
o
r
ts
o
n
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
0
8
8
-
8708
I
n
t J
E
lec
&
C
o
m
p
E
n
g
,
Vo
l.
11
,
No
.
2
,
A
p
r
il 2
0
2
1
:
1
2
5
9
-
1264
1260
ch
o
o
s
in
g
a
s
tr
o
n
g
cla
s
s
i
f
icati
o
n
alg
o
r
ith
m
.
Fo
r
ex
a
m
p
le,
i
n
[
3
]
,
a
s
u
p
p
o
r
t
v
ec
to
r
m
ac
h
in
e
(
SVM)
b
ased
r
etr
iev
al
m
eth
o
d
w
a
s
u
s
ed
to
e
x
tr
ac
t t
h
e
attr
ib
u
te
an
d
f
r
o
m
th
e
2
-
clas
s
m
u
l
ti s
u
p
p
o
r
t v
ec
to
r
m
ac
h
in
e
(
M
SVM)
an
d
K
-
n
ea
r
est
n
ei
g
h
b
o
r
in
g
.
I
n
r
esear
ch
s
u
ch
a
s
[
4
]
,
th
e
w
av
elet
p
ac
k
et
en
tr
o
p
y
a
n
d
f
ea
t
u
r
es
s
u
c
h
a
s
Fre
u
n
d
Fre
q
u
en
c
y
,
J
itter
,
Sch
i
m
er
,
M
el
f
r
eq
u
e
n
c
y
ce
p
s
tr
al
co
e
f
f
ic
ie
n
t
,
etc.
h
a
v
e
b
ee
n
u
s
ed
,
an
d
t
h
e
d
ata
class
i
f
i
ca
tio
n
h
as
b
ee
n
m
ad
e
u
s
i
n
g
t
h
e
S
V
M
.
Sp
ee
ch
r
ec
o
g
n
i
tio
n
s
y
s
te
m
s
ca
n
b
e
d
iv
id
ed
in
to
t
w
o
e
s
s
e
n
tial
p
ar
ts
o
f
t
h
e
f
ea
t
u
r
e
ex
tr
ac
tio
n
an
d
cla
s
s
i
f
i
ca
tio
n
.
Fi
g
u
r
e
1
s
h
o
w
s
t
h
e
b
lo
ck
d
iag
r
a
m
o
f
a
s
p
ee
ch
r
ec
o
g
n
itio
n
s
y
s
te
m
.
I
n
t
h
e
r
e
m
ain
d
er
o
f
t
h
is
p
ap
er
,
s
ec
tio
n
2
d
is
c
u
s
s
es
th
e
f
u
s
io
n
o
f
f
ea
tu
r
es,
f
ea
tu
r
e
d
e
g
r
ad
atio
n
w
it
h
t
h
e
m
ai
n
co
m
p
o
n
e
n
t
s
(
P
C
A
)
a
n
d
th
e
c
lass
i
f
icatio
n
s
u
s
ed
in
th
e
p
r
o
p
o
s
ed
m
eth
o
d
.
Sectio
n
3
p
r
esen
t
s
t
h
e
p
r
o
p
o
s
ed
m
et
h
o
d
.
S
ec
tio
n
4
d
is
cu
s
s
es
th
e
r
esu
lts
o
f
th
e
d
ia
g
n
o
s
ti
c
test
s
b
y
s
p
ee
ch
,
an
d
i
n
t
h
e
f
in
a
l
s
ec
tio
n
,
th
e
co
n
clu
s
io
n
is
d
r
a
w
n
.
Fig
u
r
e
1
.
B
lo
ck
d
iag
r
a
m
g
e
n
er
al
d
iag
n
o
s
i
s
o
f
s
p
ee
c
h
2.
ST
E
P
S T
O
D
E
T
E
C
T
F
E
E
L
I
NG
S B
Y
SPE
E
CH
As
s
h
o
w
n
i
n
F
ig
u
r
e
1
,
t
h
e
s
e
n
s
o
r
y
r
ec
o
g
n
itio
n
s
y
s
te
m
co
n
s
i
s
ts
o
f
f
o
u
r
p
ar
ts
.
I
n
itial
l
y
,
a
s
p
ee
ch
s
ig
n
a
l
en
ter
s
t
h
e
s
y
s
te
m
.
T
h
en
ex
tr
ac
tio
n
an
d
r
ed
u
ctio
n
f
u
n
ct
io
n
s
ar
e
p
er
f
o
r
m
ed
.
2
.
1
.
Da
t
a
ba
s
e
W
e
u
s
ed
th
r
ee
d
atab
ases
in
Ger
m
an
a
n
d
E
n
g
l
is
h
to
ev
al
u
ate
th
e
p
r
o
p
o
s
ed
m
et
h
o
d
.
T
h
e
Ger
m
a
n
B
er
lin
d
atab
ase
i
s
p
u
b
lic
l
y
a
v
ailab
le
o
n
th
e
I
n
ter
n
et,
an
d
m
an
y
r
e
s
ea
r
ch
er
s
h
a
v
e
u
s
ed
it
i
n
t
h
eir
w
o
r
k
.
T
h
is
d
atab
ase
co
n
tain
s
5
3
6
s
en
te
n
ce
s
(
1
0
s
p
ea
k
er
s
,
7
f
ee
li
n
g
s
,
1
0
w
o
r
d
s
in
cl
u
d
in
g
n
at
u
r
al
e
m
o
tio
n
s
,
B
o
r
ed
o
m
.
d
is
g
u
s
t,
Fear
,
h
ap
p
in
es
s
,
s
ad
n
ess
,
An
g
er
)
.
T
h
e
SA
VE
E
d
atab
as
e
is
in
E
n
g
lis
h
an
d
co
n
tai
n
s
4
8
0
s
en
ten
ce
s
in
7
s
en
s
es
(
n
at
u
r
al
e
m
o
tio
n
s
,
d
i
s
g
u
s
t
,
Fear
,
h
ap
p
in
e
s
s
,
An
g
er
,
s
ad
n
ess
,
s
u
r
p
r
is
e)
.
I
n
o
r
d
er
to
ev
alu
ate
th
e
p
r
o
p
o
s
ed
s
y
s
te
m
i
n
all
ex
p
er
i
m
en
ts
u
s
i
n
g
th
e
N
-
f
o
ld
cr
o
s
s
-
v
alid
atio
n
w
a
s
Used
[
2
,
4
]
.
2.
2
.
E
x
t
ra
ct
io
n
a
nd
re
du
ct
io
n o
f
f
ea
t
ures
I
n
th
i
s
p
ap
er
,
to
w
f
ea
tu
r
e
g
r
o
u
p
s
h
a
v
e
b
ee
n
in
v
e
s
ti
g
ated
.
T
h
e
f
ir
s
t
g
r
o
u
p
is
b
ased
o
n
f
u
n
d
a
m
en
tal
fr
eq
u
en
c
y
(
F0
)
,
en
er
g
y
(
E
)
,
ze
r
o
-
cr
o
s
s
in
g
r
ate
(
Z
C
R
)
(
FEZ
)
.
T
h
e
s
ec
o
n
d
g
r
o
u
p
is
b
ased
o
n
th
e
Fo
u
r
ier
p
ar
am
eter
(
FP
)
m
o
d
el.
2.
2
.
1
.
F
E
Z
co
nti
nu
o
us
f
ea
t
ur
es
FEZ
f
ea
tu
r
es
ar
e
b
ased
o
n
t
h
r
ee
f
u
n
d
a
m
e
n
tal
f
u
n
ctio
n
al
g
r
o
u
p
s
(
F0
)
,
e
n
er
g
y
(
E
)
,
a
n
d
ze
r
o
-
cr
o
s
s
i
n
g
r
ate
(
Z
C
R
)
.
T
h
ese
f
ea
t
u
r
es
ar
e
am
o
n
g
t
h
e
m
o
s
t
co
m
m
o
n
l
y
u
s
ed
f
ea
tu
r
es
i
n
r
ec
o
g
n
iz
i
n
g
s
e
n
s
a
tio
n
f
r
o
m
s
p
ee
ch
,
as
p
ar
t
o
f
th
e
s
ta
n
d
ar
d
f
ea
t
u
r
es
o
f
t
h
is
f
ield
[
5
,
6
]
.
T
h
e
Sp
ee
ch
B
ase
Fre
q
u
en
c
y
(
F0
)
is
an
i
m
p
o
r
tan
t
f
ea
t
u
r
e
th
at
p
r
o
v
id
es
th
e
to
n
e
al
an
d
r
h
y
t
h
m
ic
p
r
o
p
er
ties
o
f
th
e
s
p
ee
c
h
.
E
n
er
g
y
s
tr
o
n
g
l
y
s
ig
n
al
s
th
e
s
p
ee
ch
s
ig
n
al,
s
h
o
w
i
n
g
a
p
a
u
s
e
an
d
e
m
p
h
asi
s
o
n
s
p
ee
ch
,
an
d
i
s
h
i
g
h
l
y
d
ep
en
d
en
t
o
n
s
p
ee
ch
d
ialec
t.
T
h
e
ze
r
o
cr
o
s
s
in
g
r
ate
(
Z
C
R
)
r
ep
r
esen
t
s
th
e
m
o
m
e
n
ts
t
h
at
th
e
ad
j
ac
en
t
s
a
m
p
les
o
f
an
ac
o
u
s
tic
s
i
g
n
al
ar
e
ch
an
g
i
n
g
th
e
s
ig
n
.
R
esear
c
h
h
as
s
h
o
w
n
th
at
th
e
s
et
o
f
F0
,
e
n
er
g
y
,
a
n
d
Z
C
R
f
ea
t
u
r
es
ar
e
b
etter
th
a
n
f
ea
t
u
r
es
s
u
c
h
a
s
Fo
r
m
en
t
a
n
d
L
i
n
ea
r
p
r
ed
ictiv
e
ce
p
s
tr
al
co
ef
f
icien
ts
(
FP
C
)
[
7
,
8
]
.
I
n
th
i
s
s
t
u
d
y
,
FEZ
f
ea
tu
r
es
ar
e
e
x
tr
ac
ted
f
r
o
m
t
h
e
f
u
n
d
a
m
en
ta
l
f
r
eq
u
e
n
c
y
(
F0
)
,
en
er
g
y
(
E
)
,
ze
r
o
-
cr
o
s
s
in
g
r
ate
(
Z
C
R
)
s
i
g
n
als.
F
o
r
th
is
p
u
r
p
o
s
e,
t
h
e
s
p
ee
ch
s
ig
n
al
is
f
ir
s
t
d
i
v
id
ed
i
n
to
f
r
a
m
e
s
o
f
2
0
m
illi
s
ec
o
n
d
s
w
it
h
an
o
v
er
lap
o
f
1
0
m
illi
s
ec
o
n
d
s
.
F
u
n
d
a
m
e
n
tal
f
r
eq
u
en
c
y
(
F0
)
,
en
er
g
y
(
E
)
,
an
d
ze
r
o
-
cr
o
s
s
i
n
g
r
ate
(
Z
C
R
)
ar
e
ca
lcu
lated
f
o
r
ea
ch
f
r
a
m
e.
So
,
th
er
e
ar
e
t
h
r
ee
cu
r
v
e
s
f
o
r
f
u
n
d
a
m
en
tal
f
r
eq
u
en
c
y
(
F0
)
,
en
er
g
y
(
E
)
,
an
d
z
er
o
-
cr
o
s
s
in
g
r
ate
(
Z
C
R
)
.
Mi
n
i
m
u
m
,
m
a
x
i
m
u
m
,
av
er
ag
e,
m
ea
n
,
a
n
d
s
tan
d
ar
d
d
ev
iatio
n
o
f
th
e
s
e
cu
r
v
es
w
il
l b
e
u
s
ed
as FE
Z
attr
ib
u
te
s
.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
n
t J
E
lec
&
C
o
m
p
E
n
g
I
SS
N:
2
0
8
8
-
8708
S
p
ee
ch
emo
tio
n
r
ec
o
g
n
itio
n
b
a
s
ed
o
n
S
V
M a
n
d
K
N
N
cla
s
s
if
ica
tio
n
s
fu
s
io
n
(
Mo
h
a
mme
d
J
a
w
a
d
A
l
Du
ja
ili
)
1261
2.
2
.
2
.
F
o
urier
pa
ra
m
et
er
(
F
P
)
Fo
u
r
ier
s
er
ies
a
n
al
y
s
is
is
o
n
e
o
f
t
h
e
m
o
s
t
i
m
p
o
r
ta
n
t
a
n
a
l
y
tical
m
e
th
o
d
s
in
en
g
i
n
ee
r
in
g
.
Fo
u
r
ier
an
al
y
s
is
h
as
b
ee
n
w
id
el
y
u
s
e
d
f
o
r
s
i
g
n
al
p
r
o
ce
s
s
i
n
g
,
i
n
cl
u
d
in
g
f
ilter
in
g
,
co
r
r
elatio
n
,
en
co
d
in
g
,
i
n
te
g
r
ati
n
g
,
an
d
ex
tr
ac
ti
n
g
f
ea
t
u
r
es
f
o
r
p
atter
n
r
ec
o
g
n
itio
n
.
T
h
is
Fo
u
r
ier
an
al
y
s
is
i
s
a
s
i
g
n
al
to
d
ec
o
m
p
o
s
e
i
n
to
it
s
s
in
u
s
o
id
al
v
ib
r
atio
n
s
.
Fro
m
t
h
is
p
er
s
p
ec
tiv
e,
a
s
p
ee
c
h
s
ig
n
al
ca
n
b
e
r
eg
ar
d
ed
as
a
r
es
u
lt
o
f
a
n
e
x
cita
tio
n
s
ig
n
al
(
s
o
n
ic
a
u
d
io
o
u
tp
u
t)
p
ass
i
n
g
t
h
r
o
u
g
h
a
ti
m
e
-
v
ar
y
i
n
g
lin
ea
r
ti
m
e
f
ilter
(
p
h
o
n
o
lo
g
ical
d
u
ct
ass
e
m
b
l
y
)
th
at
m
o
d
els
t
h
e
r
eso
n
a
n
ce
p
r
o
p
er
ties
o
f
th
e
au
d
io
d
ev
ice.
A
s
p
ee
ch
s
i
g
n
al
(
n
)
x
d
iv
id
ed
in
to
L
f
r
a
m
e
ca
n
b
e
r
ep
r
esen
ted
b
y
t
h
e
co
m
b
i
n
atio
n
o
f
a
FP
m
o
d
el
in
t
h
e
f
o
r
m
(
1
)
[
9
]
:
X
(
n
)
=
∑
(
H
K
1
(
)
(
c
os
(
2
2
)
)
+
1
)
M
K
=
1
(1
)
w
h
er
e:
Fs
i
s
th
e
s
a
m
p
li
n
g
f
r
eq
u
en
c
y
o
f
th
e
s
p
ee
ch
s
i
g
n
a
l
(
n
)
x
,
an
d
th
e
a
m
p
lit
u
d
e
a
n
d
th
e
p
h
ase
is
t
h
e
h
ar
m
o
n
ic
s
i
n
u
s
o
id
al
co
m
p
o
n
en
t
k
,
l
r
ep
r
esen
ts
t
h
e
f
r
a
m
e,
an
d
M
is
th
e
n
u
m
b
er
o
f
h
ar
m
o
n
ic
ele
m
e
n
ts
o
f
s
p
ee
ch
.
T
h
e
h
ar
m
o
n
ics
o
f
t
h
i
s
m
o
d
el
ar
e
a
Fo
u
r
ier
s
er
ial
d
is
p
lay
o
f
th
e
al
ter
n
ati
n
g
co
m
p
o
n
en
t
s
o
f
t
h
e
s
p
ee
c
h
s
ig
n
al.
Fo
r
ea
ch
f
r
a
m
e
o
b
tain
ed
f
r
o
m
s
o
u
n
d
,
w
e
o
b
tain
t
h
e
v
alu
e
s
o
f
h
ar
m
o
n
ic
co
e
f
f
ici
en
ts
.
T
h
e
f
ir
s
t
1
2
0
co
ef
f
icie
n
t
s
w
ill
b
e
ch
o
s
e
n
i
n
th
e
s
e
co
ef
f
icie
n
ts
.
T
h
ese
1
2
0
co
ef
f
icien
t
s
w
it
h
f
ir
s
t
a
n
d
s
ec
o
n
d
d
eg
r
ee
s
d
if
f
er
e
n
tial
v
alu
e
s
w
ill
f
o
r
m
an
ele
m
e
n
tal
ele
m
e
n
t
ar
r
ay
o
f
3
6
0
.
T
h
e
m
a
x
i
m
u
m
,
m
in
i
m
u
m
,
a
v
er
ag
e,
m
o
d
e,
an
d
s
tan
d
ar
d
d
ev
iatio
n
o
f
an
1
8
0
0
elem
en
t a
r
r
a
y
w
ill b
e
f
o
r
m
ed
w
h
e
n
p
lacin
g
t
h
o
s
e
to
g
et
h
er
[
1
0
]
.
2.
3
.
Cla
s
s
if
ica
t
io
n
I
n
t
h
is
w
o
r
k
,
w
e
u
s
e
a
co
m
p
o
s
ite
co
n
s
tr
u
ct
b
ased
o
n
th
e
t
wo
ca
teg
o
r
ies
o
f
b
ac
k
u
p
v
ec
to
r
m
ac
h
i
n
es
(
SVM)
,
K
-
n
ea
r
e
s
t n
ei
g
h
b
o
r
(
KNN)
.
2.
3
.
1
.
Su
pp
o
rt
v
ec
t
o
r
m
a
chi
ne
(
SVM
)
Su
p
p
o
r
t
v
ec
to
r
m
ac
h
in
e
(
S
VM
)
is
o
n
e
o
f
t
h
e
s
u
p
er
v
i
s
o
r
y
lear
n
in
g
m
et
h
o
d
s
t
h
at
it
u
s
ed
f
o
r
class
i
f
icatio
n
a
n
d
r
eg
r
es
s
io
n
[
1
1
]
.
T
h
e
b
ac
k
u
p
m
ac
h
i
n
e
ca
te
g
o
r
izatio
n
i
s
a
li
n
ea
r
d
ata
ca
te
g
o
r
izatio
n
t
h
at
tr
ies
to
s
elec
t
a
lin
e
th
at
h
as
t
h
e
h
i
g
h
est
m
ar
g
i
n
o
f
co
n
f
id
en
ce
.
A
lt
h
o
u
g
h
tr
ai
n
i
n
g
at
t
h
e
f
as
test
S
VM
s
is
v
er
y
s
lo
w
,
th
is
c
lass
if
icatio
n
h
as
a
v
er
y
h
ig
h
d
e
g
r
ee
o
f
ac
cu
r
ac
y
.
I
t
h
as
b
ee
n
w
id
el
y
u
s
ed
to
r
ec
o
g
n
ize
t
h
e
f
ee
lin
g
o
f
s
p
ee
ch
[1
2,
13
].
2.
3
.
2
.
K
-
nea
re
s
t
neig
hb
o
r's a
lg
o
rit
h
m
(
K
NN)
T
h
e
K
-
n
ea
r
est
n
ei
g
h
b
o
r
(
KNN)
alg
o
r
ith
m
is
a
clas
s
i
f
icatio
n
tech
n
iq
u
e
t
h
at
is
b
ased
o
n
c
o
m
p
ar
ati
v
e
lear
n
in
g
[
14,
15
]
.
Fo
r
an
ex
p
er
i
m
en
tal
d
ata,
t
h
e
al
g
o
r
ith
m
lo
o
k
s
f
o
r
k
s
a
m
p
le
s
f
r
o
m
t
h
e
n
ea
r
est
s
a
m
p
le
s
.
I
n
th
is
m
et
h
o
d
,
it
is
d
ec
id
ed
w
h
i
ch
n
e
w
cla
s
s
to
f
all
i
n
to
;
w
h
i
ch
clas
s
if
ie
s
t
h
e
n
e
w
in
s
ta
n
ce
b
elo
n
g
s
to
a
clas
s
th
at
h
a
s
t
h
e
m
o
s
t
v
o
tes i
n
t
h
e
K
n
ea
r
est
n
eig
h
b
o
r
s
[
1
6
].
2.
3
.
3
.
M
a
j
o
rit
y
v
o
t
ing
rule:
M
VR
I
n
o
r
d
er
to
b
e
ab
le
to
u
s
e
th
e
MV
R
r
u
le,
t
h
e
m
aj
o
r
it
y
v
o
te
m
u
s
t
al
w
a
y
s
b
e
co
r
r
ec
t
[1
7
]
.
T
h
is
r
u
le
s
tates
t
h
at
t
h
e
i
n
p
u
t
o
f
x
b
e
lo
n
g
s
to
t
h
e
clas
s
i
i
f
an
d
o
n
l
y
if
t
h
e
ex
is
ti
n
g
N
cla
s
s
i
f
ier
e
x
i
s
ts
,
t
h
e
m
aj
o
r
ity
o
f
th
e
m
h
a
v
e
ch
o
s
en
i
cla
s
s
[
1
8,
19
]
.
T
o
im
p
le
m
e
n
t
t
h
is
m
et
h
o
d
,
ea
ch
lev
el
w
o
r
d
is
f
ir
s
tl
y
d
eter
m
i
n
ed
,
th
en
it
i
s
an
n
o
u
n
ce
d
a
m
o
n
g
th
e
f
o
u
r
ex
i
s
tin
g
le
v
els o
f
th
e
w
i
n
n
i
n
g
s
u
r
f
ac
e,
w
h
ic
h
h
as a
m
aj
o
r
it
y
o
f
5
w
o
r
d
s
s
elec
ted
.
3.
CO
M
P
O
NE
NT
ANA
L
YS
I
S
AL
G
O
RIHM
T
h
e
P
C
A
tech
n
iq
u
e
is
a
s
u
i
ta
b
le
m
et
h
o
d
f
o
r
r
ed
u
ci
n
g
d
ata
d
i
m
en
s
io
n
s
li
n
ea
r
l
y
.
B
y
e
li
m
i
n
ati
n
g
t
h
e
m
i
n
u
s
cu
le
co
ef
f
ic
ien
t
s
o
b
tain
ed
f
r
o
m
t
h
is
co
n
v
er
s
io
n
,
th
e
lo
s
t
d
ata
ar
e
less
th
a
n
o
th
e
r
m
et
h
o
d
s
.
I
n
th
i
s
m
et
h
o
d
,
th
e
ax
e
s
o
f
t
h
e
n
e
w
c
o
o
r
d
in
ates f
o
r
th
e
d
ata
ar
e
d
ef
i
n
ed
an
d
th
e
d
ata
i
s
e
x
p
r
ess
ed
b
ased
o
n
th
ese
a
x
e
s
o
f
th
e
n
e
w
co
o
r
d
in
ates.
T
h
e
f
i
r
s
t a
x
i
s
s
h
o
u
ld
b
e
in
th
e
d
ir
ec
tio
n
w
h
er
e
th
e
d
ata
v
ar
ia
n
ce
i
s
m
ax
i
m
ized
(
th
at
is
,
in
t
h
e
d
ir
ec
tio
n
i
n
w
h
ich
th
e
d
ata
is
s
ca
tter
ed
)
.
T
h
e
s
ec
o
n
d
a
x
is
m
u
s
t
b
e
p
er
p
en
d
icu
lar
to
t
h
e
f
ir
s
t
a
x
is
,
s
o
th
at
th
e
v
ar
ian
ce
o
f
t
h
e
d
ata
i
s
m
a
x
i
m
ized
.
Si
m
i
lar
l
y
,
t
h
e
la
ter
al
a
x
is
is
p
er
p
en
d
ic
u
lar
to
all
o
f
t
h
e
p
r
ev
io
u
s
a
x
e
s
s
o
th
at
t
h
e
d
ata
is
th
e
m
o
s
t d
is
p
er
s
e
d
in
th
a
t d
ir
ec
tio
n
[
2
0
,
2
1
]
.
4.
T
H
E
P
RO
P
O
SE
D
M
E
T
H
O
D
As
m
e
n
tio
n
ed
,
e
x
tr
ac
ti
n
g
t
h
e
p
r
o
p
er
f
ea
tu
r
e
o
f
s
p
ee
ch
,
a
n
d
an
e
f
f
icie
n
t
a
n
d
o
p
ti
m
al
cl
ass
i
f
icatio
n
s
y
s
te
m
co
n
s
i
s
ti
n
g
o
f
t
h
e
co
m
b
in
atio
n
o
f
s
e
v
er
al
cla
s
s
i
f
icatio
n
al
g
o
r
ith
m
s
p
la
y
a
ce
n
tr
al
r
o
le
in
t
h
e
f
u
n
c
tio
n
o
f
a
s
en
s
o
r
y
r
ec
o
g
n
i
tio
n
s
y
s
te
m
f
r
o
m
s
p
ee
ch
.
I
n
th
e
m
et
h
o
d
p
r
o
p
o
s
ed
in
th
i
s
p
ap
er
,
af
ter
ex
tr
ac
ti
n
g
th
e
p
r
o
p
er
ties
(
FEZ
)
,
(
FP
)
th
e
p
r
in
cip
al
co
m
p
o
n
e
n
t
an
al
y
s
is
(
P
C
A
)
alg
o
r
it
h
m
is
u
s
ed
as
a
f
ea
tu
r
e
r
ed
u
ctio
n
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
0
8
8
-
8708
I
n
t J
E
lec
&
C
o
m
p
E
n
g
,
Vo
l.
11
,
No
.
2
,
A
p
r
il 2
0
2
1
:
1
2
5
9
-
1264
1262
f
ea
t
u
r
e,
an
d
t
h
e
n
f
r
o
m
t
h
e
co
m
b
in
at
io
n
o
f
su
p
p
o
r
t
v
ec
to
r
m
ac
h
in
e
(
SVM)
a
n
d
K
-
clo
s
e
s
t
n
eig
h
b
o
r
(
KNN)
h
av
e
b
ee
n
u
s
ed
as
a
clas
s
i
f
ic
atio
n
.
A
p
p
r
o
p
r
iate
m
et
h
o
d
s
a
r
e
u
s
ed
to
in
cr
ea
s
e
t
h
e
r
eliab
i
lit
y
o
f
th
e
e
m
o
tio
n
d
etec
tio
n
s
y
s
te
m
u
s
i
n
g
s
er
ial
f
u
s
io
n
o
f
cla
s
s
i
f
ier
s
b
y
m
aj
o
r
ity
v
o
ti
n
g
la
w
.
T
o
d
ay
,
th
e
m
u
lti
p
le
clas
s
i
f
ier
s
y
s
te
m
s
(
M
C
S)
cla
s
s
i
f
ier
i
s
u
s
ed
to
r
ep
lace
t
h
e
v
er
y
co
m
p
lex
clas
s
es
t
h
at
r
eq
u
ir
e
a
lo
t
o
f
ed
u
ca
tio
n
al
ca
lcu
latio
n
s
[
2
2
]
.
T
h
er
e
ar
e
th
r
ee
m
eth
o
d
s
to
co
m
b
i
n
e
cla
s
s
if
ica
tio
n
s
:
co
m
b
i
n
atio
n
,
s
er
ia
l
an
d
p
ar
allel.
I
n
a
h
y
b
r
id
m
et
h
o
d
,
in
th
i
s
m
et
h
o
d
,
th
e
class
i
f
icatio
n
is
r
eg
u
lated
in
a
tr
ee
s
tr
u
ctu
r
e
th
at
i
s
a
co
llectio
n
o
f
d
if
f
er
en
t
class
es
[
2
3
]
.
I
n
p
ar
allel,
in
th
is
m
et
h
o
d
,
ea
ch
class
if
icat
io
n
is
ap
p
lied
in
d
ep
en
d
en
tl
y
an
d
th
e
d
ec
is
io
n
alg
o
r
ith
m
is
ap
p
lied
to
t
h
e
o
u
tp
u
t
s
.
I
n
a
s
er
ial
m
et
h
o
d
,
th
e
n
u
m
b
er
o
f
d
i
f
f
er
e
n
t
c
lass
e
s
f
o
r
o
th
er
clas
s
es
is
r
ed
u
ce
d
.
T
h
e
p
r
o
p
o
s
ed
m
et
h
o
d
w
as
co
m
p
ar
ed
to
co
n
v
e
n
tio
n
al
m
et
h
o
d
s
f
o
r
p
er
f
o
r
m
a
n
ce
ac
c
u
r
ac
y
an
d
co
m
p
u
tatio
n
al
co
m
p
le
x
it
y
[
2
4
,
2
5
]
.
5.
E
VA
L
UA
T
I
O
N
O
F
RE
S
UL
T
S
I
n
th
i
s
p
ap
er
,
w
e
h
a
v
e
ex
tr
a
cted
th
e
ch
ar
ac
ter
is
t
ics
o
f
F
E
Z
an
d
FP
f
r
o
m
t
h
e
Ger
m
a
n
d
atab
ase
s
en
te
n
ce
s
[
2
]
.
B
er
lin
's
Ger
m
a
n
d
atab
ase
i
s
p
u
b
lic
l
y
av
ailab
l
e
th
r
o
u
g
h
t
h
e
I
n
ter
n
et
a
n
d
m
a
n
y
r
esear
ch
er
s
h
a
v
e
u
s
ed
it
i
n
t
h
eir
w
o
r
k
.
T
h
is
d
at
ab
ase
co
n
tain
s
5
3
6
s
en
te
n
ce
s
(
1
0
s
p
ea
k
er
s
,
7
f
ee
lin
g
s
o
f
1
0
s
p
ee
ch
es
i
n
cl
u
d
in
g
n
atu
r
al
e
m
o
tio
n
s
,
B
o
r
ed
o
m
.
d
is
g
u
s
t,
Fear
,
h
ap
p
in
e
s
s
,
s
ad
n
ess
,
An
g
er
)
.
T
h
e
S
A
VE
E
d
at
ab
ase
is
i
n
E
n
g
l
is
h
an
d
co
n
tain
s
4
8
0
s
en
ten
ce
s
i
n
s
ev
e
n
f
ee
li
n
g
s
(
n
at
u
r
al
e
m
o
tio
n
s
,
d
is
g
u
s
t,
Fear
,
h
ap
p
in
e
s
s
,
An
g
er
,
s
ad
n
ess
,
s
u
r
p
r
is
e)
.
I
n
o
r
d
er
to
ev
alu
a
te
th
e
p
r
o
p
o
s
ed
s
y
s
te
m
,
1
0
-
f
o
ld
-
cr
o
s
s
-
v
alid
atio
n
tec
h
n
iq
u
e
w
as
u
s
ed
in
all
ex
p
er
i
m
e
n
ts
.
I
n
itiall
y
,
w
e
co
n
s
id
er
ed
th
e
f
u
s
io
n
o
f
t
h
e
f
ea
t
u
r
es
to
g
et
h
er
a
n
d
th
e
n
ap
p
lied
to
th
e
SVM
a
n
d
KNN
clas
s
i
f
icatio
n
s
,
an
d
t
h
e
r
esu
lt
s
o
f
t
h
e
d
etec
tio
n
r
ate
f
o
r
ea
ch
d
atab
ase
ac
co
r
d
in
g
t
o
th
e
clas
s
if
ier
,
in
o
r
d
er
to
b
etter
co
m
p
ar
e
th
e
d
atab
ase
o
f
Ger
m
a
n
,
E
n
g
li
s
h
w
it
h
class
T
h
e
SVM,
KNN
clau
s
es
s
h
o
w
th
e
av
er
ag
e
d
etec
t
io
n
r
ate
u
s
i
n
g
th
e
f
u
s
io
n
o
f
al
l
f
ea
t
u
r
es
a
f
ter
ap
p
ly
in
g
th
e
P
C
A
to
id
en
ti
f
y
t
h
e
s
en
s
es
i
n
T
ab
le
s
1
an
d
2
.
T
h
e
p
r
o
p
er
ty
o
f
th
e
P
C
A
i
s
t
h
at
it
d
o
es
n
o
t
d
elete
an
y
i
m
p
o
r
tan
t
f
ea
t
u
r
e,
as
w
e
ca
n
s
ee
,
th
e
b
est
r
esu
lt
s
f
o
r
t
h
e
Ger
m
a
n
d
a
tab
ase
in
t
h
e
SVM
cla
s
s
i
f
icat
i
o
n
u
s
in
g
FE
Z
f
ea
tu
r
e
i
n
T
ab
le
3
f
o
r
s
ev
en
s
en
s
es
w
it
h
an
ac
cu
r
ac
y
o
f
8
5
.
1
%
an
d
an
ex
ec
u
tio
n
ti
m
e
o
f
0
.
0
5
s
ec
o
n
d
s
.
An
d
th
e
b
est
r
es
u
lt
s
f
o
r
th
e
Ger
m
a
n
d
atab
ase
an
d
KNN
class
i
f
icat
io
n
ar
e
th
e
f
u
s
io
n
FP
+FE
Z
f
ea
tu
r
es
p
r
o
p
er
ties
w
it
h
a
p
r
ec
is
io
n
d
etec
tio
n
o
f
8
7
.
8
5
%
an
d
a
r
u
n
ti
m
e
ex
ec
u
t
io
n
ti
m
e
o
f
0
.
4
8
s
ec
o
n
d
s
,
as
s
h
o
w
n
in
T
ab
le
4
.
T
h
en
,
w
e
p
er
f
o
r
m
ed
s
i
m
ilar
ex
p
er
i
m
e
n
ts
o
n
t
h
e
Ger
m
a
n
l
an
g
u
a
g
e
d
atab
ase
an
d
t
h
e
cla
s
s
if
ica
tio
n
o
f
SVM,
KNN
i
n
t
h
e
E
n
g
lis
h
d
atab
ase
f
o
r
th
e
clas
s
i
f
icatio
n
o
f
SVM
an
d
KN
N,
an
d
th
e
r
e
s
u
l
ts
ar
e
p
r
esen
ted
i
n
T
ab
les
5
an
d
6
a
s
s
ee
n
i
n
t
h
e
s
e
tab
les.
T
h
e
h
i
g
h
e
s
t
d
etec
tio
n
r
ate
f
o
r
t
h
e
E
n
g
li
s
h
d
ata
b
ase
is
f
r
o
m
t
h
e
f
u
s
io
n
o
f
FE
Z
p
r
o
p
er
ties
in
SVM
class
i
f
icatio
n
w
i
th
a
r
eso
l
u
tio
n
o
f
8
5
.
2
%
an
d
a
r
u
n
ti
m
e
e
x
ec
u
t
io
n
ti
m
e
o
f
0
.
0
5
s
ec
o
n
d
s
,
an
d
t
h
e
h
i
g
h
est
d
etec
tio
n
r
ate
f
o
r
th
e
E
n
g
lis
h
la
n
g
u
a
g
e
d
atab
ase
i
n
K
N
N
class
if
ica
tio
n
b
y
f
u
s
io
n
o
f
all
f
ea
tu
r
e
s
w
it
h
a
d
etec
tio
n
ac
cu
r
ac
y
o
f
9
0
.
8
3
%
An
d
th
e
e
x
ec
u
t
io
n
ti
m
e
o
f
t
h
e
alg
o
r
ith
m
is
0
.
3
5
s
ec
o
n
d
s
.
T
ab
le
1
.
C
o
m
p
ar
is
o
n
o
f
th
e
r
at
e
o
f
d
iag
n
o
s
i
s
o
f
e
m
o
tio
n
s
in
t
h
e
Ger
m
a
n
d
atab
ase
w
ith
al
l th
e
class
i
f
icatio
n
s
G
e
r
man
+
K
N
N
G
e
r
man
+
S
V
M
d
a
t
a
b
a
se
,
c
l
a
ssi
f
i
e
r
s
T
i
me
S
e
c
A
c
c
u
.
%
T
i
me
S
e
c
A
c
c
u
%.
f
e
a
t
u
r
e
s
FP
4
9
.
2
0
.
7
8
7
.
6
6
0
.
4
5
F
E
Z
8
5
.
1
0
.
0
5
8
7
.
1
1
0
.
0
1
F
P
+
F
EZ
84
1
.
0
6
8
7
.
8
5
0
.
4
8
T
ab
le
2
.
C
o
m
p
ar
is
o
n
o
f
th
e
r
at
e
o
f
e
m
o
tio
n
d
etec
tio
n
in
t
h
e
E
n
g
li
s
h
d
atab
ase
w
ith
al
l th
e
class
i
f
icatio
n
s
En
g
l
i
sh
+
K
N
N
En
g
l
i
sh
+
S
V
M
d
a
t
a
b
a
se
,
c
l
a
ssi
f
i
e
r
s
T
i
me
S
e
c
A
c
c
u
.
%
T
i
me
S
e
c
A
c
c
u
%.
f
e
a
t
u
r
e
s
FP
5
3
.
1
0
.
6
5
9
1
.
0
4
0
.
3
7
F
E
Z
8
5
.
2
0
.
0
5
8
7
.
2
9
0
.
0
1
F
P
+
F
EZ
85
1
.
0
6
9
0
.
8
3
0
.
3
5
T
ab
le
3
.
I
n
ter
ac
tio
n
m
a
tr
ix
t
h
e
b
est r
esu
lt o
f
t
h
e
Ger
m
an
d
atab
ase
w
it
h
cla
s
s
i
f
i
ca
tio
n
SVM
F
E
Z
A
c
c
u
r
a
c
=
8
5
.
1
%,
T
i
me
=
0
.
0
5
S
e
c
N
e
u
t
.
B
o
r
e
d
D
i
sg
.
F
e
a
h
a
p
S
a
d
.
A
n
g
.
N
e
u
t
.
9
7
.
6
0
0
0
0
.
7
0
1
.
6
B
o
r
e
d
.
8
.
6
4
9
0
.
1
1
.
2
0
0
0
0
D
i
sg
.
8
.
6
9
0
89
0
2
.
1
0
0
F
e
a
.
2
1
.
7
1
.
4
4
0
71
0
0
5
.
7
H
p
p
.
1
9
.
7
0
0
0
80
0
0
S
a
d
.
9
.
6
7
1
.
6
1
0
1
.
6
0
8
5
.
4
1
.
6
A
n
g
.
1
2
.
6
0
0
1
.
2
1
.
2
1
.
2
6
83
N
e
u
t
r
a
l
=
N
e
u
t
,
B
o
r
e
d
o
m
=
B
o
r
e
d
,
D
i
sg
=
d
i
sg
u
st
,
F
e
a
=
F
e
a
r
,
h
a
p
=
h
a
p
p
i
n
e
ss,
sa
d
=
s
a
d
n
e
ss,
a
n
g
=
A
n
g
e
r
T
ab
le
4
.
I
n
ter
ac
tio
n
m
a
tr
ix
t
h
e
b
est r
esu
lt o
f
t
h
e
Ger
m
an
d
atab
ase
w
it
h
t
h
e
clas
s
if
ica
tio
n
o
f
K
NN
F
E
Z
A
c
c
u
r
a
c
y
=
8
5
.
2
%
,
T
i
me
=
0
.
0
5
S
e
c
N
e
u
t
.
D
i
sg
.
F
e
a
H
a
p
A
n
g
.
S
a
d
.
S
u
r
p
.
N
e
u
t
.
98
0
0
0
1
.
9
0
0
D
i
sg
.
0
82
0
0
19
0
0
F
e
a
r
0
0
77
0
25
0
0
H
a
p
0
0
0
85
15
0
0
A
n
g
.
17
0
0
0
84
0
0
S
a
d
.
0
0
0
0
20
80
0
S
u
r
p
.
0
0
0
0
22
0
7
8
.
4
Evaluation Warning : The document was created with Spire.PDF for Python.
I
n
t J
E
lec
&
C
o
m
p
E
n
g
I
SS
N:
2
0
8
8
-
8708
S
p
ee
ch
emo
tio
n
r
ec
o
g
n
itio
n
b
a
s
ed
o
n
S
V
M a
n
d
K
N
N
cla
s
s
if
ica
tio
n
s
fu
s
io
n
(
Mo
h
a
mme
d
J
a
w
a
d
A
l
Du
ja
ili
)
1263
T
ab
le
5
.
I
n
ter
ac
tio
n
m
a
tr
ix
t
h
e
b
est r
esu
lt o
f
t
h
e
E
n
g
l
is
h
d
atab
ase
w
ith
t
h
e
SV
M
class
i
f
icatio
n
F
P
+
F
EZ
A
c
c
u
r
a
c
y
=
8
7
.
8
5
%
,
T
i
me
=
0
.
4
8
S
e
c
N
e
u
t
.
B
o
r
e
d
D
i
sg
.
F
e
a
h
a
p
S
a
d
.
A
n
g
.
N
e
u
t
.
9
2
.
2
0
0
.
9
3
3
.
4
0
0
.
8
B
o
r
e
d
.
0
8
8
.
9
1
.
2
3
0
1
.
3
6
.
2
D
i
sg
.
0
0
95
0
2
.
2
2
.
8
0
F
e
a
.
0
0
7
.
3
77
7
.
3
2
.
9
5
.
6
H
p
p
.
9
.
9
0
1
.
5
2
85
0
2
.
9
S
a
d
.
0
4
.
8
3
0
1
.
7
0
92
1
.
7
A
n
g
.
0
5
.
0
6
0
1
.
3
0
1
.
3
9
3
N
e
u
t
r
a
l
=
N
e
u
,
d
i
sg
u
st
=
d
i
sg
,
F
e
a
r
=
F
e
a
,
h
a
p
p
i
n
e
ss =
h
a
p
,
A
n
g
e
r
=
A
n
g
,
sad
=
s
a
d
n
e
ss,
s
u
r
p
r
i
se
=
su
r
p
T
ab
le
6
.
I
n
ter
ac
tio
n
m
atr
i
x
t
h
e
b
est r
esu
lt o
f
t
h
e
E
n
g
l
is
h
d
atab
ase
w
ith
t
h
e
KN
N
class
i
f
icatio
n
F
P
+
F
EZ
A
c
c
u
r
a
c
y
=
9
0
.
8
3
%,
T
i
me
=
0
.
3
5
S
e
c
N
e
u
t
.
D
i
sg
.
F
e
a
H
a
p
A
n
g
.
S
a
d
.
S
u
r
p
.
N
e
u
t
.
97
0
.
9
0
.
8
0
0
0
.
8
0
D
i
sg
.
0
88
3
.
4
0
5
3
.
5
0
F
e
a
r
0
3
.
3
89
0
0
1
.
6
6
.
6
6
H
a
p
3
.
4
1
.
7
1
.
7
93
0
0
1
.
7
A
n
g
.
5
0
3
.
4
2
89
0
1
.
7
S
a
d
.
0
3
.
4
1
.
7
0
3
.
3
92
0
S
u
r
p
.
0
1
.
7
5
0
1
.
6
0
9
0
.
7
6.
CO
NCLU
SI
O
N
T
o
s
o
lv
e
th
e
p
r
o
b
lem
o
f
d
etec
tin
g
t
h
e
d
ir
ec
t
f
ee
li
n
g
s
o
f
th
e
s
p
ea
k
er
,
a
s
en
s
atio
n
m
o
d
el
is
p
r
esen
ted
to
class
i
f
y
s
ev
e
n
s
e
n
s
e
s
.
I
n
t
h
i
s
r
esear
ch
,
t
w
o
f
ea
t
u
r
e
g
r
o
u
p
s
h
av
e
b
ee
n
i
n
v
esti
g
ated
.
T
h
e
f
ir
s
t
g
r
o
u
p
is
b
ased
o
n
f
u
n
d
a
m
e
n
tal
f
r
eq
u
e
n
c
y
(
F
0
)
,
en
er
g
y
(
E
)
,
ze
r
o
-
cr
o
s
s
i
n
g
r
ate
(
Z
C
R
)
(
FE
Z
)
.
T
h
e
s
ec
o
n
d
g
r
o
u
p
is
b
ased
o
n
th
e
Fo
u
r
ier
p
ar
a
m
eter
(
FP
)
m
o
d
el.
T
h
ese
f
ea
tu
r
es
ar
e
al
s
o
co
n
s
id
er
ed
as
th
e
in
p
u
t
p
ar
a
m
eter
f
o
r
s
u
p
p
o
r
t
v
ec
to
r
m
a
ch
i
n
e,
t
h
e
clo
s
est
KNN
n
ei
g
h
b
o
r
.
I
n
o
r
d
er
t
o
ev
al
u
ate
t
h
e
p
r
o
p
o
s
ed
s
y
s
te
m
,
t
h
e
p
r
in
cip
al
co
m
p
o
n
e
n
t
a
n
al
y
s
is
(
P
C
A
)
is
p
r
o
p
o
s
e
d
to
r
ed
u
ce
th
e
f
ea
t
u
r
e
an
d
S
VM
,
KNN
clo
s
est
n
eig
h
b
o
r
.
Fo
r
th
i
s
p
u
r
p
o
s
e,
co
m
p
ar
ati
v
e
te
s
ts
,
i
n
clu
d
i
n
g
t
h
e
f
u
s
io
n
o
f
th
e
ch
ar
ac
ter
is
tics
to
g
et
h
er
,
an
d
t
h
e
n
t
h
e
f
u
s
io
n
o
f
th
e
ch
ar
ac
ter
is
tic
s
is
ta
k
e
n
u
s
in
g
P
C
A
.
E
x
p
er
i
m
e
n
tal
r
es
u
lts
h
a
v
e
p
r
o
v
en
g
o
o
d
r
esu
lt
s
i
n
f
u
s
io
n
o
f
f
ea
tu
r
e
s
an
d
th
e
b
es
t
r
es
u
lt
s
w
i
th
t
h
e
Ger
m
an
d
atab
ase
o
n
f
u
s
io
n
o
f
all
f
ea
tu
r
es
a
n
d
P
C
A
w
it
h
KNN
c
lass
i
f
ica
tio
n
w
i
th
a
r
eso
lu
tio
n
o
f
8
7
.
8
5
% a
n
d
ti
m
i
n
g
o
f
ex
ec
u
tio
n
o
f
0
.
4
8
s
ec
.
T
h
e
b
est r
es
u
lt
s
w
i
th
th
e
E
n
g
li
s
h
d
atab
ase
o
n
f
u
s
io
n
o
f
all
f
ea
t
u
r
es a
n
d
P
C
A
w
i
th
KNN
class
if
ica
tio
n
w
it
h
a
r
es
o
lu
tio
n
o
f
9
0
.
8
3
% a
n
d
r
u
n
ti
m
e
o
f
0
.
3
5
s
ec
o
n
d
s
.
RE
F
E
R
E
NC
E
S
[1
]
D
.
J
.
F
r
a
n
c
e
,
R
.
G
.
S
h
i
a
v
i
,
S
.
S
i
l
v
e
r
m
a
n
,
M
.
S
i
l
v
e
r
m
a
n
,
a
n
d
M
.
W
i
l
k
e
s
,
"
A
c
o
u
s
t
i
c
a
l
p
r
o
p
e
r
t
i
e
s
o
f
s
p
e
e
c
h
a
s
in
d
ica
to
rs
o
f
d
e
p
re
ss
io
n
a
n
d
s
u
icid
a
l
r
is
k
,
"
IEE
E
T
ra
n
sa
c
ti
o
n
s o
n
B
io
-
me
d
ica
l
E
n
g
i
n
e
e
rin
g
,
v
o
l.
4
7
,
v
o
l.
7
,
p
p
.
8
2
9
-
8
3
7
,
2
0
0
7
.
[2
]
M
.
G
ri
m
m
,
K.
Kro
sc
h
e
l,
a
n
d
S
.
Na
ra
y
a
n
a
n
,
"
T
h
e
V
e
ra
Am
M
it
tag
G
e
r
m
a
n
a
u
d
io
-
v
isu
a
l
e
m
o
ti
o
n
a
l
sp
e
e
c
h
d
a
tab
a
se
,
"
2
0
0
8
IEE
E
In
ter
n
a
ti
o
n
a
l
Co
n
fer
e
n
c
e
o
n
M
u
lt
ime
d
i
a
a
n
d
Exp
o
,
Ha
n
n
o
v
e
r
,
2
0
0
8
,
p
p
.
8
6
5
-
8
6
8
.
[3
]
Ke
rk
e
n
i,
L
e
il
a
,
e
t
a
l.
,
"
A
u
to
m
a
t
ic
S
p
e
e
c
h
Em
o
ti
o
n
Re
c
o
g
n
it
i
o
n
Us
in
g
M
a
c
h
in
e
L
e
a
rn
in
g
,
"
S
o
c
ia
l
M
e
d
ia
a
n
d
M
a
c
h
in
e
L
e
a
r
n
in
g
.
I
n
tec
h
Op
e
n
,
p
p
.
1
-
1
6
,
2
0
1
9
.
[4
]
El
y
a
d
e
r
a
n
i,
M
.
K.,
M
a
h
m
o
o
d
ia
n
,
S
.
H.,
a
n
d
S
h
e
ik
h
i,
G
.
,
"
Wav
e
let
P
a
c
k
e
t
En
tro
p
y
in
S
p
e
a
k
e
r
-
In
d
e
p
e
n
d
e
n
t
Em
o
ti
o
n
a
l
S
tat
e
De
tec
ti
o
n
f
ro
m
S
p
e
e
c
h
S
ig
n
a
l,
"
J
o
u
rn
a
l
o
f
In
telli
g
e
n
t
Pro
c
e
d
u
re
s
i
n
El
e
c
trica
l
T
e
c
h
n
o
lo
g
y
,
v
o
l.
5
,
n
o
.
2
0
,
p
p
.
6
7
-
7
4
,
2
0
1
5
.
[5
]
F
u
,
L
i
q
in
,
Xia
M
a
o
,
a
n
d
L
ij
ian
g
Ch
e
n
,
"
S
p
e
a
k
e
r
in
d
e
p
e
n
d
e
n
t
e
m
o
ti
o
n
re
c
o
g
n
it
i
o
n
b
a
se
d
o
n
S
V
M
/HM
M
s
f
u
sio
n
s
y
ste
m
,
"
2
0
0
8
In
ter
n
a
ti
o
n
a
l
C
o
n
f
e
re
n
c
e
o
n
Au
d
io
,
L
a
n
g
u
a
g
e
a
n
d
I
ma
g
e
Pro
c
e
ss
in
g
,
S
h
a
n
g
h
a
i
,
2
0
0
8
,
p
p
.
6
1
-
65
.
[6
]
L
in
,
Y.
L
.
,
a
n
d
W
e
i,
G
.
,
"
S
p
e
e
c
h
e
m
o
ti
o
n
re
c
o
g
n
it
io
n
b
a
se
d
o
n
H
M
M
a
n
d
S
V
M
,
"
2
0
0
5
In
ter
n
a
ti
o
n
a
l
C
o
n
fer
e
n
c
e
o
n
M
a
c
h
in
e
L
e
a
rn
i
n
g
a
n
d
Cy
b
e
rn
e
ti
c
s,
Gu
a
n
g
zh
o
u
,
Ch
i
n
a
, v
o
l.
8
,
2
0
0
5
,
p
p
.
4
8
9
8
-
4
9
0
1
.
[7
]
R.
Co
w
ie,
E.
Do
u
g
las
-
Co
w
ie,
N.
T
s
a
p
a
tso
u
li
s,
S
.
Ko
ll
ias
,
W
.
F
e
ll
e
n
z
,
J.
,
Ta
y
lo
r,
"
E
m
o
ti
o
n
re
c
o
g
n
it
io
n
i
n
h
u
m
a
n
-
c
o
m
p
u
ter i
n
tera
c
ti
o
n
,
"
IEE
E
S
ig
n
a
l
Pro
c
e
ss
in
g
M
a
g
a
zin
e
,
v
o
l.
1
8
,
n
o
.
1
,
p
p
.
3
2
-
80
,
2
0
0
1
.
[8
]
M
.
El
Ay
a
d
i,
M
.
S
.
Ka
m
e
l,
a
n
d
F
.
Ka
rra
y
,
"
S
u
rv
e
y
o
n
sp
e
e
c
h
e
m
o
ti
o
n
re
c
o
g
n
it
io
n
:
F
e
a
tu
re
s
,
c
las
sif
i
c
a
ti
o
n
sc
h
e
m
e
s,
a
n
d
d
a
tab
a
se
s,"
Pa
tt
e
rn
Rec
o
g
n
it
io
n
,
v
o
l.
4
4
,
n
o
.
3
,
p
p
.
5
7
2
–
5
8
7
,
2
0
1
1
[9
]
J.
-
H.
Ye
h
,
T
.
-
L
.
P
a
o
,
C.
-
Y.
L
i
n
,
Y.
-
W
.
T
s
a
i,
a
n
d
Y.
-
T
.
Ch
e
n
,
"
S
e
g
m
e
n
t
-
b
a
se
d
e
m
o
ti
o
n
re
c
o
g
n
it
io
n
f
ro
m
c
o
n
t
in
u
o
u
s
M
a
n
d
a
rin
Ch
i
n
e
se
sp
e
e
c
h
,
"
Co
mp
u
ter
s in
Hu
m
a
n
Beh
a
v
i
o
r
,
v
o
l
.
2
7
,
n
o
.
5
,
p
p
.
1
5
4
5
-
1
5
5
2
,
2
0
1
1
.
[1
0
]
J.
S
.
W
a
lk
e
r,
“
F
o
u
rier
se
ries
,
”
in
En
c
y
c
lo
p
e
d
i
a
o
f
Ph
y
sic
a
l
S
c
i
e
n
c
e
a
n
d
T
e
c
h
n
o
lo
g
y
.
Ne
w
Yo
rk
,
NY
,
USA
:
A
c
a
d
e
m
ic,
2
0
0
1
.
[1
1
]
R.
M
c
A
u
la
y
a
n
d
T
.
Qu
a
ti
e
ri,
"
S
p
e
e
c
h
a
n
a
l
y
sis/S
y
n
th
e
sis
b
a
se
d
o
n
a
sin
u
so
id
a
l
re
p
re
se
n
ta
ti
o
n
,
"
in
IEE
E
T
ra
n
sa
c
ti
o
n
s
o
n
Aco
u
s
ti
c
s,
S
p
e
e
c
h
,
a
n
d
S
i
g
n
a
l
Pr
o
c
e
ss
in
g
,
v
o
l.
3
4
,
n
o
.
4
,
p
p
.
7
4
4
-
7
5
4
,
1
9
8
6
.
[1
2
]
A
lt
u
n
,
Ha
li
s,
a
n
d
G
ö
k
h
a
n
P
o
lat,
"
Bo
o
stin
g
se
lec
ti
o
n
o
f
sp
e
e
c
h
re
late
d
f
e
a
tu
re
s
to
i
m
p
ro
v
e
p
e
rf
o
r
m
a
n
c
e
o
f
m
u
lt
i
-
c
las
s S
V
M
s i
n
e
m
o
ti
o
n
d
e
tec
ti
o
n
,
"
Exp
e
rt S
y
ste
ms
wit
h
Ap
p
li
c
a
ti
o
n
s
,
v
o
l.
3
6
,
n
o
.
4
,
p
p
.
8
1
9
7
-
8
2
0
3
,
2
0
0
9
.
[1
3
]
B.
S
c
h
u
ll
e
r,
G
.
Rig
o
ll
,
M
.
L
a
n
g
,
“
S
p
e
e
c
h
e
m
o
ti
o
n
re
c
o
g
n
it
i
o
n
c
o
m
b
in
in
g
a
c
o
u
stic
f
e
a
tu
re
s
a
n
d
li
n
g
u
isti
c
in
f
o
rm
a
ti
o
n
in
a
h
y
b
rid
su
p
p
o
rt
v
e
c
to
r
m
a
c
h
in
e
-
b
e
li
e
f
n
e
tw
o
rk
a
r
c
h
it
e
c
tu
re
,
”
2
0
0
4
IEE
E
I
n
ter
n
a
ti
o
n
a
l
C
o
n
fer
e
n
c
e
o
n
Ac
o
u
stics
,
S
p
e
e
c
h
,
a
n
d
S
ig
n
a
l
Pro
c
e
ss
in
g
,
M
o
n
trea
l,
Qu
e
,
v
o
l
.
1
,
2
0
0
4
,
p
p
.
5
7
7
-
5
8
0
.
[1
4
]
L
e
e
,
Ch
u
l
M
in
,
a
n
d
S
h
rik
a
n
th
S
.
Na
ra
y
a
n
a
n
,
"
T
o
wa
rd
d
e
tec
ti
n
g
e
m
o
ti
o
n
s
in
sp
o
k
e
n
d
ial
o
g
s
,
"
IEE
E
T
ra
n
sa
c
ti
o
n
s
o
n
S
p
e
e
c
h
a
n
d
A
u
d
i
o
Pro
c
e
ss
in
g
,
v
o
l.
1
3
,
n
o
.
2
,
p
p
.
2
9
3
-
3
0
3
,
2
0
0
5
.
[1
5
]
L
a
n
je
w
a
r,
Ra
h
u
l
B.
,
S
w
a
ru
p
M
a
th
u
rk
a
r,
a
n
d
Ni
les
h
P
a
tel,
"
Im
p
le
m
e
n
tatio
n
a
n
d
c
o
m
p
a
riso
n
o
f
sp
e
e
c
h
e
m
o
ti
o
n
re
c
o
g
n
it
io
n
sy
ste
m
u
sin
g
g
a
u
ss
i
a
n
m
ix
tu
re
m
o
d
e
l
(g
m
m
)
a
n
d
k
-
n
e
a
re
st
n
e
ig
h
b
o
r
(k
-
n
n
)
tec
h
n
iq
u
e
s,
"
Pro
c
e
d
i
a
c
o
mp
u
ter
sc
ien
c
e
,
v
o
l.
4
9
,
p
p
.
50
-
57
,
2
0
1
5
.
[1
6
]
Zh
o
u
,
Jia
n
,
e
t
a
l
.
,
"
S
p
e
e
c
h
e
m
o
ti
o
n
re
c
o
g
n
i
ti
o
n
b
a
se
d
o
n
r
o
u
g
h
se
t
a
n
d
S
V
M
,
"
2
0
0
6
5
t
h
IEE
E
In
ter
n
a
t
io
n
a
l
Co
n
fer
e
n
c
e
o
n
Co
g
n
i
ti
v
e
In
f
o
rm
a
ti
c
s
,
Be
ij
in
g
,
2
0
0
6
,
p
p
.
5
3
-
61
.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
0
8
8
-
8708
I
n
t J
E
lec
&
C
o
m
p
E
n
g
,
Vo
l.
11
,
No
.
2
,
A
p
r
il 2
0
2
1
:
1
2
5
9
-
1264
1264
[1
7
]
M
.
G
ri
m
m
,
K.
Kro
sc
h
e
l,
E.
M
o
we
r,
a
n
d
S
.
Na
ra
y
a
n
a
n
,
“
P
rim
it
iv
e
s
-
b
a
se
d
e
v
a
lu
a
ti
o
n
a
n
d
e
st
ima
ti
o
n
o
f
e
m
o
ti
o
n
s
in
sp
e
e
c
h
,
”
S
p
e
e
c
h
Co
mm
u
n
ica
ti
o
n
,
v
o
l.
4
9
,
n
o
.
1
0
-
1
1
,
p
p
.
7
8
7
-
8
0
0
,
2
0
0
7
.
[1
8
]
S
h
e
ik
h
a
n
,
M
a
n
s
o
u
r,
M
a
h
d
i
Be
jan
i,
a
n
d
Da
v
o
o
d
G
h
a
ra
v
ian
,
"
M
o
d
u
lar
n
e
u
ra
l
-
S
VM
sc
h
e
m
e
f
o
r
s
p
e
e
c
h
e
m
o
ti
o
n
re
c
o
g
n
it
io
n
u
sin
g
A
NO
V
A
f
e
a
t
u
re
se
lec
ti
o
n
m
e
th
o
d
,
"
Ne
u
ra
l
Co
m
p
u
t
in
g
a
n
d
Ap
p
li
c
a
t
io
n
s
,
v
o
l.
2
3
,
n
o
.
1
,
p
p
.
2
1
5
-
2
2
7
,
2
0
1
3
.
[1
9
]
G
ian
n
a
k
o
p
o
u
l
o
s,
G
e
o
rg
io
s,
e
t
a
l.
,
"
En
se
m
b
le
m
a
jo
rit
y
v
o
ti
n
g
c
las
si
f
ier
f
o
r
sp
e
e
c
h
e
m
o
ti
o
n
r
e
c
o
g
n
it
io
n
a
n
d
p
re
d
ictio
n
,
"
J
o
u
r
n
a
l
o
f
S
y
st
e
ms
a
n
d
I
n
fo
rm
a
ti
o
n
T
e
c
h
n
o
l
o
g
y
,
v
o
l.
1
6
,
n
o
.
3
,
p
p
.
2
2
2
-
2
3
2
,
2
0
1
4
.
[2
0
]
Jo
ll
if
f
e
,
I.
T
.
,
"
M
a
th
e
m
a
ti
c
a
l
a
n
d
sta
ti
stica
l
p
ro
p
e
rti
e
s
o
f
p
o
p
u
lati
o
n
p
r
in
c
ip
a
l
c
o
m
p
o
n
e
n
ts,
"
Prin
c
i
p
a
l
C
o
mp
o
n
e
n
t
An
a
lys
is
,
p
p
.
8
-
22
,
2
0
0
2
.
[2
1
]
S
h
a
sh
id
h
a
r
G
.
Ko
o
lag
u
d
i,
K
a
n
d
S
re
e
n
iv
a
sa
R.
,
"
Em
o
ti
o
n
re
c
o
g
n
it
io
n
f
ro
m
sp
e
e
c
h
:
a
re
v
iew
,
"
In
ter
n
a
ti
o
n
a
l
J
o
u
rn
a
l
o
f
S
p
e
e
c
h
T
e
c
h
n
o
l
o
g
y
,
v
o
l.
1
5
,
p
p
.
9
9
-
1
1
7
,
2
0
1
2
.
[2
2
]
P
.
L
a
u
k
k
a
,
D.
Ne
ib
e
rg
,
M
.
F
o
rse
ll
,
I.
Ka
rlsso
n
a
n
d
K.
El
e
n
iu
s
“
Ex
p
re
ss
io
n
o
f
a
ff
e
c
t
in
sp
o
n
t
a
n
e
o
u
s
sp
e
e
c
h
:
A
c
o
u
stic
c
o
rre
late
s
a
n
d
a
u
to
m
a
ti
c
d
e
tec
ti
o
n
o
f
irri
tatio
n
a
n
d
re
sig
n
a
ti
o
n
,
”
Co
mp
u
ter
S
p
e
e
c
h
a
n
d
L
a
n
g
u
a
g
e
,
v
o
l
.
2
5
,
n
o
.
1
,
p
p
.
84
-
1
0
4
,
2
0
1
1
.
[2
3
]
Oli
F
.
,
“
F
u
sio
n
o
f
m
u
lt
ip
le p
a
tt
e
rn
c
las
sif
iers
,
”
J
o
u
rn
a
l
Arti
c
le
,
2
0
0
3
[2
4
]
Bu
ss
o
,
Ca
rlo
s,
S
u
n
g
b
o
k
L
e
e
,
a
n
d
S
h
r
ik
a
n
th
Na
ra
y
a
n
a
n
,
"
A
n
a
l
y
sis
o
f
e
m
o
ti
o
n
a
ll
y
sa
li
e
n
t
a
sp
e
c
ts
o
f
f
u
n
d
a
m
e
n
tal
f
re
q
u
e
n
c
y
f
o
r
e
m
o
ti
o
n
d
e
tec
ti
o
n
,
"
I
EE
E
T
ra
n
sa
c
ti
o
n
s
o
n
a
u
d
i
o
,
s
p
e
e
c
h
,
a
n
d
la
n
g
u
a
g
e
p
ro
c
e
ss
in
g
,
v
o
l.
1
7
,
n
o
.
4
,
p
p
.
5
8
2
-
596
,
2
0
0
9
.
[2
5
]
Os
ter,
A
.
M
.
,
a
n
d
A
rn
e
Risb
e
rg
,
"
T
h
e
id
e
n
ti
f
ica
ti
o
n
o
f
th
e
m
o
o
d
o
f
a
sp
e
a
k
e
r
b
y
h
e
a
rin
g
i
m
p
a
ired
l
isten
e
rs,
"
S
L
T
-
Qu
a
rte
rly
Pro
g
re
ss
S
t
a
t
u
s R
e
p
o
rt
,
v
o
l.
2
7
,
n
o
.
4
,
p
p
.
7
9
-
9
0
,
1
9
8
6
.
B
I
O
G
RAP
H
I
E
S
O
F
AUTH
O
RS
M
o
h
a
m
m
e
d
J
a
w
a
d
Al
-
D
u
ja
il
i
Al
-
K
h
a
z
r
a
ji
a
w
a
rd
e
d
B.
S
.
d
e
g
re
e
in
c
o
m
m
u
n
ica
ti
o
n
e
n
g
in
e
e
ri
n
g
f
ro
m
Un
iv
e
rsit
y
o
f
A
l
-
F
u
ra
t
A
l
-
Aw
s
a
t
Tec
h
n
ica
l,
T
e
c
h
n
ica
l
Co
ll
e
g
e
o
f
En
g
in
e
e
rin
g
,
Na
jaf
,
Ira
q
in
2
0
0
8
a
n
d
M
.
S
.
d
e
g
re
e
i
n
c
o
m
m
u
n
ica
ti
o
n
s
y
ste
m
e
n
g
in
e
e
rin
g
f
ro
m
F
e
rd
o
w
si
u
n
iv
e
rsit
y
,
Ira
n
,
in
2
0
1
8
.
C
u
rre
n
tl
y
,
h
e
is a
m
e
m
b
e
r
s
taff
a
t
th
e
D
e
p
a
rt
m
e
n
t
o
f
El
e
c
tro
n
ic an
d
Co
m
m
u
n
ica
ti
o
n
,
F
a
c
u
lt
y
o
f
En
g
in
e
e
rin
g
,
Un
iv
e
rsity
o
f
Ku
f
a
,
Ira
q
.
His
re
se
a
r
c
h
in
tere
st
in
c
lu
d
e
s
th
e
d
e
v
e
lo
p
m
e
n
t
o
f
W
irele
ss
c
o
m
m
u
n
ica
ti
o
n
s
a
n
d
sig
n
a
l
p
ro
c
e
ss
in
g
a
s
w
e
ll
a
s
i
m
a
g
e
,
sp
e
e
c
h
p
ro
c
e
ss
in
g
a
n
d
ra
d
a
r
,
5
G
.
Ema
il
:
M
o
h
a
m
m
e
d
.
c
h
a
ll
a
b
@u
o
k
u
f
a
.
e
d
u
.
i
q
Abb
a
s
E
b
r
a
h
i
m
i
-
M
o
g
h
a
d
a
m
re
c
e
iv
e
d
th
e
B.
S
c
.
d
e
g
re
e
in
e
lec
tri
c
a
l
e
n
g
in
e
e
rin
g
f
ro
m
th
e
S
h
a
rif
Un
iv
e
rsit
y
o
f
T
e
c
h
n
o
lo
g
y
,
T
e
h
ra
n
,
Ira
n
,
in
1
9
9
1
,
th
e
M
.
S
c
.
d
e
g
re
e
in
e
lec
tri
c
a
l
e
n
g
in
e
e
rin
g
f
ro
m
th
e
K.
N.
T
o
o
si
Un
iv
e
rsity
o
f
Te
c
h
n
o
l
o
g
y
,
Teh
ra
n
,
in
1
9
9
5
,
a
n
d
t
h
e
P
h
.
D.
d
e
g
re
e
in
e
lec
tri
c
a
l
a
n
d
c
o
m
p
u
ter
e
n
g
in
e
e
rin
g
f
o
r
m
M
c
M
a
ste
r
Un
iv
e
rsit
y
,
Ha
m
il
to
n
,
ON
,
Ca
n
a
d
a
.
He
h
a
s
b
e
e
n
a
n
A
s
sista
n
t
P
r
o
f
e
ss
o
r
w
it
h
th
e
El
e
c
tri
c
a
l
En
g
in
e
e
rin
g
De
p
a
rtm
e
n
t,
F
e
rd
o
w
si
Un
iv
e
rsit
y
o
f
M
a
sh
h
a
d
,
M
a
sh
h
a
d
,
Ira
n
,
si
n
c
e
2
0
1
1
.
Em
a
il
:
a
.
e
b
ra
h
im
i.
m
@
g
m
a
il
.
c
o
m
Ahm
e
d
Fa
tla
w
i
a
w
a
r
d
e
d
B.
S
.
d
e
g
re
e
in
te
c
h
n
o
l
o
g
ie
s
e
n
g
in
e
e
ri
n
g
o
f
c
o
m
p
u
ter
f
ro
m
A
l
k
a
f
e
e
l
u
n
iv
e
rsity
,
N
a
ja
f
,
Ir
a
q
in
2
0
1
5
a
n
d
M
.
S
.
d
e
g
re
e
in
c
o
m
p
u
ter
e
n
g
i
n
e
e
rin
g
/
a
rti
f
icia
l
in
telli
g
e
n
c
e
a
n
d
ro
b
o
ti
c
s
f
ro
m
F
e
rd
o
w
si
Un
iv
e
rsit
y
o
f
M
a
sh
h
a
d
,
M
a
sh
h
a
d
,
Ira
n
,
i
n
2
0
1
8
.
C
u
rre
n
t
ly
,
h
e
is
a
stu
d
e
n
t
sta
ff
P
h.
D
.
Em
a
il
:ah
.
f
a
tl
a
w
i@
m
a
i
l.
u
m
.
a
c
.
ir
Evaluation Warning : The document was created with Spire.PDF for Python.