IA
E
S
In
t
e
r
n
a
t
i
o
n
a
l
J
o
u
r
n
a
l
o
f
Ar
t
i
f
i
c
i
a
l
I
n
t
e
l
l
i
g
e
n
c
e
(I
J
-
AI
)
Vo
l
.
8
, N
o
.
1
,
Ma
r
c
h
201
9
, p
p
.
7
~
13
IS
S
N
:
2252
-
8938
,
DOI
:
1
0
.
1
1
5
9
1
/
i
j
a
i
.
v
8
.i
1
.p
p
7
-
13
r
7
Jou
r
n
al
h
om
e
page
:
ht
t
p:
/
/
i
ae
s
co
r
e
.c
o
m
/o
n
lin
e
/in
d
e
x
.p
h
p
/I
J
A
I
Spe
e
c
h
r
e
c
o
g
ni
t
i
o
n
o
f
m
o
r
o
c
c
a
n
di
a
l
e
c
t
us
i
ng
hi
dde
n
ma
r
k
o
v
mo
d
e
l
s
Be
z
o
u
i
Mo
u
a
z
,
Be
n
i
-
hs
s
a
ne
A
bde
r
r
a
hi
m
,
El
m
o
u
t
a
o
u
a
k
k
i
l
A
bde
l
m
a
j
i
d
De
p
a
r
t
me
n
t
of
Co
m
p
u
t
e
r
Sc
i
e
n
c
e
s
,
Fa
c
u
l
t
y
o
f
Sc
i
e
n
c
e
s
,
C
houa
i
b
D
oukka
l
i
Un
i
v
e
r
s
i
t
y
, E
lja
d
id
a
, M
o
r
o
r
c
c
o
Ar
t
i
c
l
e
I
n
f
o
AB
S
T
RACT
Ar
t
i
c
l
e
h
i
s
t
o
r
y
:
Re
c
e
i
v
e
d
D
e
c
2
7
, 2
0
1
8
Re
v
i
s
e
d
Fe
b
18
, 2
0
1
9
Ac
c
e
p
t
e
d
Fe
b
26,
2019
Th
i
s
p
a
p
e
r
a
d
d
r
e
s
s
e
s
t
h
e
d
e
v
e
l
o
p
m
e
n
t
o
f
a
n
A
u
t
o
m
a
t
i
c
Sp
e
e
c
h
R
e
c
o
g
n
i
t
i
o
n
(A
S
R
)
s
y
s
t
e
m
fo
r
t
h
e
M
o
ro
c
c
a
n
D
i
a
l
e
c
t
.
D
i
a
l
e
c
t
a
l
A
ra
b
i
c
(D
A
)
re
fe
rs
t
o
t
h
e
da
y
-
to
-
da
y
ve
r
na
c
ul
a
r
s
s
poke
n
i
n
t
he
A
r
a
b
w
or
l
d.
I
n
f
a
c
t
,
M
or
oc
c
a
n
D
i
a
l
e
c
t
is
v
e
r
y
d
if
f
e
r
e
n
t
f
r
o
m
th
e
M
o
d
e
r
n
St
a
n
d
a
r
d
A
r
a
b
i
c
(
M
SA
)
b
e
c
a
u
s
e
i
t
i
s
hi
ghl
y
i
nf
l
ue
nc
e
d
by
t
he
F
r
e
nc
h
L
a
ngua
ge
.
I
t
i
s
obs
e
r
ve
d
t
hr
oughout
a
l
l
A
r
a
b
co
u
n
t
r
i
es
t
h
at
st
an
d
ar
d
A
r
ab
i
c
w
i
d
el
y
w
r
i
t
t
en
an
d
u
sed
f
o
r
o
f
f
i
ci
al
sp
eech
,
ne
w
s
pa
pe
r
s
,
publ
i
c
a
dm
i
ni
s
t
r
a
t
i
on
a
nd
s
c
hool
but
not
us
e
d
i
n
e
ve
r
y
da
y
co
n
v
er
sat
i
o
n
an
d
d
i
al
ect
i
s
w
i
d
el
y
sp
o
k
en
i
n
ev
er
y
d
ay
l
i
f
e
b
u
t
al
m
o
st
n
ev
er
wr
i
t
t
e
n
.
we
p
r
o
p
o
s
e
t
o
u
s
e
t
h
e
M
e
l
F
r
e
q
u
e
n
c
y
C
e
p
s
t
r
a
l
C
o
e
f
f
i
c
i
e
n
t
(
M
F
C
C
)
fe
a
t
u
re
s
t
o
s
p
e
c
i
fy
t
h
e
b
e
s
t
s
p
e
a
k
e
r
i
d
e
n
t
i
fi
c
a
t
i
o
n
s
y
s
t
e
m
.
T
h
e
e
x
t
ra
c
t
e
d
sp
eech
f
eat
u
r
es
ar
e
q
u
an
t
i
zed
to
a
n
u
m
b
e
r
o
f
c
e
n
tr
o
id
s
u
s
in
g
v
e
c
to
r
qua
nt
i
z
a
t
i
on
a
l
gor
i
t
hm
.
T
he
s
e
c
e
nt
r
oi
ds
c
ons
t
i
t
ut
e
t
he
c
ode
book
of
t
ha
t
sp
eak
er
.
M
F
C
C
’
s
ar
e
cal
cu
l
at
ed
i
n
t
r
ai
n
i
n
g
p
h
ase
an
d
ag
ai
n
i
n
t
est
i
n
g
p
h
ase.
Sp
e
a
k
e
r
s
u
t
t
e
r
e
d
s
a
m
e
w
o
r
d
s
o
n
c
e
i
n
a
t
r
a
i
n
i
n
g
s
e
s
s
i
o
n
a
n
d
o
n
c
e
i
n
a
te
s
tin
g
sessi
o
n
l
at
er
.
T
h
e
E
u
cl
i
d
ean
d
i
st
an
ce
b
et
w
een
t
h
e
M
F
C
C
’
s
o
f
each
sp
eak
er
i
n
tr
a
in
in
g
p
h
a
s
e
to
th
e
c
e
n
tr
o
id
s
o
f
in
d
iv
id
u
a
l
s
p
e
a
k
e
r
in
te
s
tin
g
p
h
a
s
e
is
me
a
s
u
r
e
d
a
n
d
t
h
e
s
p
e
a
k
e
r
i
s
i
d
e
n
t
i
f
i
e
d
a
c
c
o
r
d
i
n
g
t
o
t
h
e
mi
n
i
mu
m
E
u
c
l
i
d
e
a
n
di
s
t
a
nc
e
.
T
he
c
ode
i
s
d
ev
el
o
p
ed
i
n
t
h
e
M
A
T
L
A
B
en
v
i
r
o
n
m
en
t
an
d
p
er
f
o
r
m
s
th
e
id
e
n
tif
ic
a
tio
n
s
a
tis
f
a
c
to
r
ily
.
Ke
y
wo
r
d
s
:
AS
R
DA
H
MM
MF
C
C
MS
A
Co
p
y
r
i
g
h
t
©
2
0
1
9
In
s
t
i
t
u
t
e
o
f
A
d
v
a
n
c
e
d
E
n
g
i
n
e
e
r
i
n
g
a
n
d
S
c
i
e
n
c
e
.
Al
l
r
i
g
h
t
s
r
e
s
e
r
v
e
d
.
Co
r
r
e
s
p
o
n
d
i
n
g
Au
t
h
o
r
:
Be
z
o
u
i
M
o
u
a
z
,
De
p
a
r
t
m
e
n
t
o
f
Co
m
p
u
t
e
r
S
c
i
e
n
c
e
s
,
Ch
o
u
a
i
b
D
o
u
k
k
a
l
i
U
n
i
v
e
r
s
i
t
y
,
73,
L
ot
F
i
r
da
ous
s
,
A
ppt
1,
A
v
J
a
br
a
ne
K
ha
l
i
l
J
a
br
a
ne
,
E
l
j
a
di
da
24100,
M
or
oc
c
o.
Em
a
i
l
:
mb
e
z
o
u
i
@
g
ma
i
l
.
c
o
m
1.
IN
T
R
O
D
U
C
T
IO
N
Th
e
m
a
j
o
r
i
t
y
o
f
p
r
e
v
i
o
u
s
w
o
r
k
i
n
A
r
a
b
i
c
A
S
R
h
a
s
f
o
c
u
s
e
d
o
n
t
h
e
f
o
r
m
a
l
s
t
a
n
d
a
r
d
A
r
a
b
i
c
la
n
g
u
a
g
e
th
a
t
is
k
n
o
w
n
a
s
M
o
d
e
r
n
S
ta
n
d
a
r
d
A
r
a
b
ic
(
M
S
A
)
.
M
S
A
is
n
o
t
th
e
la
n
g
u
a
g
e
o
f
o
r
d
in
a
r
y
d
is
c
u
s
s
io
n
s
a
n
d
se
v
e
r
a
l
c
o
m
m
u
n
i
c
a
t
i
o
n
s
i
n
a
l
l
A
r
a
b
i
c
c
o
u
n
t
r
i
e
s.
T
h
e
p
o
p
u
l
a
t
i
o
n
u
se
o
t
h
e
r
A
r
a
b
i
c
v
a
r
i
e
t
i
e
s
i
n
e
v
e
r
y
d
a
y
l
i
f
e
th
a
t
is
k
n
o
w
n
a
s
D
ia
le
c
ta
l
A
r
a
b
ic
(
D
A
)
.
A
si
g
n
i
f
i
c
a
n
t
p
r
o
b
l
e
m
i
n
A
r
a
b
i
c
A
S
R
i
s
t
h
e
e
x
i
st
e
n
c
e
o
f
q
u
i
t
e
m
a
n
y
di
f
f
e
r
e
nt
di
a
l
e
c
t
s
e
.
g.
M
or
oc
c
a
n,
T
uni
s
i
a
n,
E
gypt
i
a
n,
S
a
udi
,
I
r
a
qi
e
t
c
.
E
ve
r
y
c
ount
r
y
ha
s
i
t
s
ow
n
di
a
l
e
c
t
,
a
nd
so
m
e
t
i
m
e
s
t
h
e
r
e
e
x
i
st
d
i
f
f
e
r
e
n
t
d
i
a
l
e
c
t
s
w
i
t
h
i
n
t
h
e
sa
m
e
c
o
u
n
t
r
y
[
1
]
.
M
o
r
e
o
v
e
r
,
th
e
d
if
f
e
r
e
n
t
A
r
a
b
ic
d
ia
le
c
ts
ar
e
o
n
l
y
s
p
o
k
en
an
d
n
o
t
f
o
r
m
al
l
y
w
r
i
t
t
en
an
d
s
i
g
n
i
f
i
can
t
,
s
y
n
t
act
i
c,
l
ex
i
cal
,
m
o
r
p
h
o
l
o
g
i
cal
,
p
h
o
n
o
l
o
g
i
cal
an
d
di
f
f
e
r
e
nc
e
s
e
xi
s
t
be
t
w
e
e
n
t
he
di
a
l
e
c
t
s
a
nd
t
he
s
t
a
nda
r
d
f
or
m
.
I
n
t
hi
s
w
or
k
w
e
pr
opos
e
t
o
us
e
t
he
M
e
l
Fr
e
q
u
e
n
c
y
C
e
p
s
tr
a
l
C
o
e
f
f
ic
ie
n
t
(
M
F
C
C
)
f
e
a
tu
r
e
s
f
o
r
d
e
s
ig
n
in
g
a
s
o
u
n
d
-
de
pe
nde
nt
a
nd
s
pe
c
i
f
y
t
he
be
s
t
sp
e
a
k
e
r
i
d
e
n
t
i
f
i
c
a
t
i
o
n
sy
st
e
m
.
T
h
e
e
x
t
r
a
c
t
e
d
sp
e
e
c
h
f
e
a
t
u
r
e
s
(
M
F
C
C
’
s)
o
f
a
sp
e
a
k
e
r
a
r
e
q
u
a
n
t
i
z
e
d
t
o
a
num
be
r
of
c
e
nt
r
oi
ds
us
i
ng
ve
c
t
or
qua
nt
i
z
a
t
i
on
a
l
gor
i
t
hm
.
T
he
s
e
c
e
nt
r
oi
ds
c
ons
t
i
t
ut
e
t
he
c
ode
book
of
t
ha
t
sp
e
a
k
e
r
[
2
]
.
M
F
C
C
’
s
a
r
e
c
a
l
c
u
l
a
t
e
d
i
n
t
r
a
i
n
i
n
g
p
h
a
se
a
n
d
a
g
a
i
n
i
n
t
e
st
i
n
g
p
h
a
se
.
S
p
e
a
k
e
r
s
u
t
t
e
r
e
d
sa
m
e
w
o
r
d
s
onc
e
i
n
a
t
r
a
i
ni
ng
s
e
s
s
i
on
a
nd
onc
e
i
n
a
t
e
s
t
i
ng
s
e
s
s
i
on
l
a
t
e
r
.
T
he
E
uc
l
i
de
a
n
di
s
t
a
nc
e
be
t
w
e
e
n
t
he
M
F
C
C
’
s
of
e
ach
s
p
eak
er
i
n
t
r
ai
n
i
n
g
p
h
as
e
t
o
t
h
e
cen
t
r
o
i
d
s
o
f
i
n
d
i
v
i
d
u
al
s
p
eak
er
i
n
t
es
t
i
n
g
p
h
as
e
i
s
m
eas
u
r
ed
an
d
t
h
e
Evaluation Warning : The document was created with Spire.PDF for Python.
r
IS
S
N
:
2252
-
8938
IJ
-
AI
Vo
l
.
8
, N
o
.
1
,
Ma
r
c
h
20
1
9
:
7
–
13
8
sp
e
a
k
e
r
i
s
i
d
e
n
t
i
f
i
e
d
a
c
c
o
r
d
i
n
g
t
o
t
h
e
m
i
n
i
m
u
m
E
u
c
l
i
d
e
a
n
d
i
st
a
n
c
e
.
T
h
e
c
o
d
e
i
s
d
e
v
e
l
o
p
e
d
i
n
t
h
e
M
A
T
L
A
B
en
v
i
r
o
n
m
en
t
an
d
p
er
f
o
r
m
s
t
h
e
i
d
en
t
i
f
i
cat
i
o
n
s
at
is
f
a
c
to
r
ily
.
2.
HI
D
D
E
N
M
A
R
KOV
M
OD
E
L
S
Hi
d
d
e
n
M
a
r
k
o
v
M
o
d
e
l
s
,
i
n
t
r
o
d
u
c
e
d
i
n
t
h
e
e
a
r
l
y
1
9
7
0
s
[3
],
b
e
c
a
m
e
t
h
e
p
e
rfe
c
t
s
o
l
u
t
i
o
n
t
o
t
h
e
pr
obl
e
m
s
of
a
ut
om
a
t
i
c
s
pe
e
c
h
r
e
c
ogni
t
i
on.
T
he
a
c
ous
t
i
c
s
i
gna
l
of
s
pe
e
c
h
i
s
m
ode
l
e
d
by
a
s
m
a
l
l
s
e
t
of
a
c
ous
t
i
c
uni
t
s
,
w
hi
c
h
c
a
n
be
co
n
s
i
d
er
ed
as
el
em
en
t
ar
y
s
o
u
n
d
s
o
f
t
h
e
l
an
g
u
ag
e.
T
r
ad
i
t
i
o
n
al
l
y
,
t
h
e
ch
o
s
en
u
n
i
t
i
s
t
h
e
Ph
o
n
e
m
e
,
t
h
e
r
e
b
y
t
h
e
w
o
r
d
i
s
f
o
r
m
e
d
b
y
c
o
n
c
a
t
e
n
a
t
i
n
g
t
h
e
m
.
Mo
r
e
s
p
e
c
i
f
i
c
u
n
i
t
s
c
a
n
b
e
u
s
e
d
a
s
s
y
l
l
a
b
l
e
s
,
di
s
yl
l
a
bl
e
s
,
phone
m
e
s
i
n
c
ont
e
xt
,
t
he
r
e
by
m
a
ki
ng
t
he
m
ode
l
m
or
e
di
s
c
r
i
m
i
na
t
i
ng,
but
t
hi
s
t
he
or
e
t
i
c
a
l
im
p
r
o
v
e
m
e
n
t
is
lim
ite
d
in
p
r
a
c
tic
e
b
y
th
e
c
o
m
p
le
x
ity
in
v
o
lv
e
d
a
n
d
e
s
tim
a
tio
n
p
r
o
b
le
m
s
.
T
h
e
s
p
e
e
c
h
s
ig
n
a
l
can
b
e
l
i
k
en
ed
t
o
a
s
er
i
es
o
f
u
n
i
t
s
.
I
n
t
h
e
co
n
t
ex
t
o
f
M
ar
k
o
v
A
S
R
,
t
h
e
aco
u
s
t
i
c
u
n
i
t
s
ar
e
m
o
d
el
ed
b
y
H
M
M
as
sh
o
w
n
in
F
ig
u
r
e
1
w
h
ic
h
a
r
e
ty
p
ic
a
lly
le
f
t
-
ri
g
h
t
t
ri
s
t
a
t
e
.
Fi
g
u
r
e
1
.
H
M
M
u
s
e
d
t
o
p
o
l
o
g
y
.
At
e
a
c
h
s
t
a
t
e
o
f
t
h
e
M
a
r
k
o
v
m
o
d
e
l
,
t
h
e
r
e
i
s
a
p
r
o
b
a
b
i
l
i
t
y
d
i
s
t
r
i
b
u
t
i
o
n
a
s
s
o
c
i
a
t
e
d
,
mo
d
e
l
i
n
g
t
h
e
ge
ne
r
a
t
i
on
of
a
c
ous
t
i
c
ve
c
t
or
s
vi
a
t
hi
s
s
t
a
t
e
.
a
n
H
M
M
i
s
c
ha
r
a
c
t
e
r
i
z
e
d
by
s
e
ve
r
a
l
pa
r
a
m
e
t
e
r
s
:
−
N:
t
h
e
n
u
m
b
e
r
o
f
t
h
e
s
t
a
t
e
s
o
f
t
h
e
m
o
d
e
l
.
-
A
=
=
is
th
e
m
a
tr
ix
o
f
tr
a
n
s
itio
n
p
r
o
b
a
b
ilitie
s
o
n
th
e
s
e
t o
f
s
ta
te
s
o
f
th
e
m
o
d
e
l.
-
B
=
(
)
=
is
th
e
m
a
tr
ix
o
f
e
m
ittin
g
p
r
o
b
a
b
ilitie
s
o
f
th
e
o
b
s
e
r
v
a
tio
n
s
fo
r
th
e
s
ta
te
.
-
π
i
s
t
he
i
ni
t
i
a
l
di
s
t
r
i
but
i
on
of
s
t
a
t
e
s
.
3.
ACO
US
T
I
C
M
O
DE
L
S
AND
P
ARAM
E
T
E
RS
Th
e
s
p
e
e
c
h
s
i
g
n
a
l
c
o
n
t
a
i
n
s
m
a
n
y
o
t
h
e
r
e
l
e
m
e
n
t
s
m
o
r
e
t
h
a
n
t
h
e
l
i
n
g
ui
s
t
i
c
m
e
s
s
a
ge
:
i
nf
or
m
a
t
i
on
re
l
a
t
e
d
t
o
t
h
e
s
p
e
a
k
e
r,
t
h
e
re
c
o
rd
i
n
g
c
o
n
d
i
t
i
o
n
s
,
e
t
c
.
.
.
In
a
d
d
i
t
i
o
n
,
t
h
e
v
a
ri
a
b
i
l
i
t
y
a
n
d
re
d
u
n
d
a
n
c
y
o
f
t
h
e
sp
e
e
c
h
si
g
n
a
l
m
a
k
e
s
i
t
d
i
f
f
i
c
u
l
t
t
o
u
se
a
s
su
c
h
.
I
t
i
s
t
h
e
r
e
f
o
r
e
n
e
c
e
ssa
r
y
t
o
e
x
t
r
a
c
t
t
h
e
p
a
r
a
m
e
t
e
r
s
t
h
a
t
a
r
e
de
pe
nde
nt
on
t
he
l
i
ngui
s
t
i
c
m
e
s
s
a
ge
.
T
he
s
e
pa
r
a
m
e
t
e
r
s
a
r
e
e
s
t
i
m
a
t
e
d
vi
a
s
l
i
di
ng
w
i
ndow
s
on
t
he
s
i
gna
l
.
T
hi
s
an
al
y
s
i
s
w
i
n
d
o
w
u
s
ed
t
o
es
t
i
m
at
e
t
h
e
s
i
g
n
al
o
n
a
s
t
at
i
o
n
ar
y
p
o
r
t
i
o
n
o
f
a
co
n
s
i
d
er
ed
s
i
g
n
al
:
t
y
p
i
cal
l
y
1
0
t
o
3
0
ms
l
i
mi
t
i
n
g
s
i
d
e
e
f
f
e
c
t
s
a
n
d
d
i
s
c
o
n
t
i
n
u
i
t
i
e
s
o
f
th
e
s
ig
n
a
l
v
ia
a
H
a
m
m
in
g
w
in
d
o
w
.
I
n
o
u
r
e
x
p
e
r
im
e
n
ts
,
w
e
u
s
e
25m
s
a
s
a
w
i
ndow
s
i
z
e
.
T
he
m
a
j
or
i
t
y
of
pa
r
a
m
e
t
e
r
s
r
e
pr
e
s
e
nt
t
he
f
r
e
que
nc
y
s
pe
c
t
r
um
a
nd
i
t
s
e
vol
ut
i
on
ove
r
a
wi
n
d
o
w
s
i
z
e
.
P
a
r
a
m
e
t
e
r
i
z
a
t
i
o
n
t
e
c
h
n
i
q
u
e
s
t
h
a
t
a
r
e
t
h
e
m
o
s
t
c
o
m
m
o
n
l
y
u
s
e
d
a
r
e
:
P
L
P
P
e
r
c
e
pt
ua
l
L
i
ne
a
r
Pr
e
d
i
c
t
i
o
n
:
s
p
e
c
t
r
a
l
d
o
m
a
i
n
,
L
PC
C
L
i
n
e
a
r
Pr
e
d
i
c
t
i
o
n
C
e
p
s
t
r
a
l
C
o
e
f
f
i
c
i
e
n
t
s
:
t
i
m
e
d
o
m
a
i
n
,
M
FC
C
M
e
l
Fr
e
q
u
e
n
c
y
C
e
p
s
t
r
a
l
c
o
e
f
f
i
c
i
e
n
t
s
:
c
e
p
s
t
r
a
l
d
o
m
a
i
n
[4
].
F
o
r
o
u
r
w
o
rk
,
w
e
h
a
v
e
u
s
e
d
M
F
C
C
p
a
ra
m
e
t
ri
z
a
t
i
o
n
fo
r
th
e
f
e
a
tu
r
e
e
x
tr
a
c
tio
n
.
O
u
r
f
ir
s
t
in
te
r
v
en
t
i
o
n
i
n
t
h
e
r
eco
g
n
i
t
i
o
n
s
y
s
t
em
i
s
i
n
t
h
e
p
h
as
e
o
f
l
ab
el
i
n
g
s
o
u
n
d
f
i
l
es
.
In
l
a
rg
e
v
o
c
a
b
u
l
a
ry
A
S
R
s
y
s
t
e
m
s
,
D
B
N
s
a
re
u
s
e
d
t
o
re
p
re
s
e
n
t
s
u
b
u
n
i
t
s
o
f
w
o
rd
s
(s
u
c
h
a
s
p
h
o
n
e
s
).
F
o
r
t
h
e
Ar
a
b
i
c
l
a
n
g
u
a
g
e
,
i
t
i
s
t
y
p
i
c
a
l
t
o
h
a
v
e
a
r
o
u
n
d
3
8
m
o
d
e
l
s
(
p
h
o
n
e
s
)
.
Th
e
e
x
a
c
t
p
hone
s
e
t
de
pe
nds
on
t
he
di
c
t
i
ona
r
y
t
ha
t
i
s
us
e
d.
W
or
d
m
ode
l
s
c
a
n
be
c
ons
t
r
uc
t
e
d
a
s
a
c
om
bi
na
t
i
on
of
t
he
s
ub
w
or
d
m
ode
l
s
.
I
n
pr
a
c
t
i
c
e
,
t
he
r
e
a
l
i
z
a
t
i
on
of
one
a
nd
t
he
s
a
m
e
phone
di
f
f
e
r
s
a
l
ot
de
pe
ndi
ng
on
i
t
s
ne
i
ghbor
i
ng
phone
s
c
a
l
l
e
d
‘p
h
o
n
e
c
o
n
t
e
x
t
’[5
].
S
p
e
ech
r
eco
g
n
i
t
i
o
n
u
s
e
co
n
t
ex
t
d
ep
en
d
s
o
n
p
h
o
n
et
i
c
al
p
h
ab
et
s
,
i
n
w
h
i
ch
t
h
er
e
ar
e
o
n
e
o
r
mo
r
e
u
n
i
t
s
f
o
r
e
a
c
h
p
h
o
n
e
me
i
n
t
h
e
c
o
n
t
e
x
t
o
f
s
u
r
r
o
u
n
d
i
n
g
p
h
o
n
e
me
s
.
S
e
v
e
r
a
l
o
f
t
h
e
mo
r
e
c
o
mmo
n
s
c
h
e
me
s
ar
e
m
o
n
o
p
h
o
n
es
,
b
i
p
h
o
n
es
an
d
t
r
i
p
h
o
n
es
.
Fi
g
u
r
e
2
sh
o
w
s
t
h
e
a
r
a
b
i
c
w
o
rd
‘
را
ز
’
:
[Z
A
R
]
i
n
a
m
o
n
o
p
h
o
n
e
,
bi
phone
a
nd
t
r
i
phone
r
e
pr
e
s
e
nt
a
t
i
on.
Evaluation Warning : The document was created with Spire.PDF for Python.
IJ
-
AI
IS
S
N
:
2252
-
8938
r
Spe
e
c
h
r
e
c
ogni
t
i
on
of
m
or
oc
c
an
di
al
e
c
t
us
i
ng
hi
dde
n
ma
r
k
o
v
mo
d
e
l
s
(
Be
z
o
u
i
M
o
u
a
z
)
9
‘s
i
l
’
re
fe
rs
t
o
t
h
e
s
i
l
e
n
c
e
a
t
t
h
e
s
t
a
rt
a
n
d
t
h
e
e
n
d
o
f
t
h
e
u
t
t
e
ra
n
c
e
,
w
h
i
c
h
is
m
o
d
e
le
d
a
s
a
‘
p
h
o
n
e
’
to
o
.
Fi
g
u
r
e
2
.
(a
)
M
o
n
o
p
h
o
n
e
,
(b
)
B
i
p
h
o
n
e
an
d
(
c)
T
r
i
p
h
o
n
e
w
i
t
h
o
n
e
h
i
d
d
en
v
ar
i
ab
l
e
(
H
M
M
)
f
o
r
t
h
e
A
r
ab
i
c
wo
r
d
‘
Z
AR
’
.
4.
SP
E
E
C
H
R
E
C
O
G
N
I
T
I
O
N
Li
k
e
a
n
y
o
t
h
e
r
p
a
t
t
e
r
n
r
e
c
o
g
n
i
t
i
o
n
s
y
s
t
e
m
s
,
t
h
e
p
r
o
c
e
s
s
o
f
p
e
r
f
o
r
m
i
n
g
s
p
e
a
k
e
r
r
e
c
o
g
n
i
t
i
o
n
c
o
n
s
i
s
t
s
on
t
w
o
pha
s
e
s
na
m
e
l
y:
t
r
a
i
ni
ng
a
nd
t
e
s
t
i
ng.
T
r
a
i
ni
ng
i
s
t
he
pr
oc
e
s
s
of
f
a
m
i
l
i
a
r
i
z
i
ng
t
he
s
ys
t
e
m
w
i
t
h
t
he
voi
c
e
ch
ar
act
er
i
s
t
i
cs
o
f
t
h
e
s
p
eak
er
s
r
eg
i
s
t
er
i
n
g
b
y
ex
t
r
act
f
eat
u
r
es
f
r
o
m
each
s
p
eak
er
[6
].
T
h
e
b
l
o
c
k
d
i
a
g
ra
m
of
tr
a
in
in
g
p
h
a
s
e
is
s
h
o
w
n
in
F
ig
u
r
e
3
.
F
e
a
tu
r
e
v
e
c
to
r
s
r
e
p
r
e
s
e
n
tin
g
th
e
v
o
ic
e
c
h
a
r
a
c
te
r
is
tic
s
o
f
th
e
s
p
eak
er
ar
e
ex
t
r
act
ed
f
r
o
m
t
h
e
t
r
ai
n
i
n
g
u
t
t
er
an
ces
an
d
ar
e
u
s
ed
f
o
r
b
u
i
l
d
i
n
g
t
h
e
r
e
fe
re
n
c
e
m
o
d
e
l
s
.
D
u
ri
n
g
t
e
s
t
i
n
g
,
si
m
i
l
a
r
f
e
a
t
u
r
e
v
e
c
t
o
r
s
a
r
e
e
x
t
r
a
c
t
e
d
f
r
o
m
t
h
e
t
e
st
u
t
t
e
r
a
n
c
e
,
a
n
d
t
h
e
d
e
g
r
e
e
o
f
t
h
e
i
r
m
a
t
c
h
w
i
t
h
t
h
e
r
e
f
e
r
e
n
c
e
is
o
b
ta
in
e
d
u
s
in
g
s
e
v
e
r
a
l
m
at
ch
i
n
g
al
g
o
r
i
t
h
m
s
.
f
eat
u
r
e
m
at
ch
i
n
g
p
r
o
ces
s
i
s
p
er
f
o
r
m
ed
t
o
d
eci
d
e
w
h
et
h
er
t
h
es
e
fe
a
t
u
re
s
b
e
l
o
n
g
t
o
a
p
re
v
i
o
u
s
l
y
k
n
o
w
n
s
p
e
a
k
e
r
p
a
t
t
e
rn
o
r
n
o
t
.
A
s
c
h
e
m
a
t
i
c
d
i
a
g
ra
m
o
f
t
h
e
t
e
s
t
i
n
g
p
h
a
s
e
as
sh
o
w
n
in
F
ig
u
r
e
4
.
F
ig
u
r
e
3
.
Th
e
b
l
o
c
k
d
i
a
g
r
a
m
o
f
t
h
e
t
r
a
i
n
i
n
g
m
o
d
e
.
Fi
g
u
r
e
4
.
Th
e
b
l
o
c
k
d
i
a
g
r
a
m
o
f
t
h
e
r
e
c
o
g
n
i
t
i
o
n
m
o
d
e
.
Th
e
p
r
o
c
e
s
s
o
f
p
e
r
f
o
r
m
i
n
g
s
p
e
a
k
e
r
i
d
e
n
t
i
f
i
c
a
t
i
o
n
c
o
n
s
i
s
t
s
o
f
t
w
o
m
o
d
e
s
:
a
t
r
a
i
n
i
n
g
m
o
d
e
a
n
d
a
re
c
o
g
ni
t
i
on
m
ode
.
I
n
t
he
t
r
a
i
ni
ng
pha
s
e
,
a
da
t
a
ba
s
e
of
s
pe
a
ke
r
’
s
pa
t
t
e
r
n
i
s
us
e
d
t
o
e
xt
r
a
c
t
f
e
a
t
ur
e
s
f
r
om
e
a
c
h
sp
e
a
k
e
r
.
T
h
e
se
f
e
a
t
u
r
e
s
a
r
e
u
se
d
t
o
t
r
a
i
n
a
n
e
u
r
a
l
n
e
t
w
o
r
k
.
I
n
t
h
e
t
e
st
i
n
g
p
h
a
se
,
f
e
a
t
u
r
e
s
a
r
e
e
x
t
r
a
c
t
e
d
f
r
o
m
ev
er
y
i
n
co
m
i
n
g
s
p
eak
er
an
d
a
f
eat
u
r
e
m
at
ch
i
n
g
p
r
o
ces
s
i
s
p
er
f
o
r
m
ed
t
o
d
eci
d
e
w
h
et
h
er
t
h
es
e
f
eat
u
r
es
b
el
o
n
g
to
a
p
r
e
v
io
u
s
ly
k
n
o
w
n
s
p
e
a
k
e
r
p
a
tte
r
n
o
r
n
o
t.
A
s
c
h
e
m
a
tic
d
ia
g
r
a
m
o
f
th
e
s
te
p
s
o
f
th
e
p
r
o
p
o
s
e
d
d
e
te
c
tio
n
sy
st
e
m
i
s
sh
o
w
n
i
n
F
i
g
ur
e
4.
Th
e
s
t
e
p
s
o
f
t
h
e
f
e
a
t
u
r
e
e
x
t
r
a
c
t
i
o
n
p
r
o
c
e
s
s
f
r
o
m
a
f
l
o
w
ch
ar
t
can
b
e
s
u
m
m
ar
i
zed
as
f
o
l
l
o
w
s
:
−
Th
e
s
p
e
e
c
h
s
i
g
n
a
l
c
a
n
b
e
u
s
e
d
i
n
t
i
m
e
d
o
m
a
i
n
o
r
i
n
a
n
o
t
h
e
r
d
i
s
c
r
e
t
e
t
r
a
n
s
f
o
r
m
d
o
m
a
i
n
.
Th
e
D
C
T,
D
S
T
an
d
D
W
T
can
b
e
u
s
ed
f
o
r
t
h
i
s
p
u
r
p
o
s
e.
−
MF
C
C
s
a
n
d
p
o
l
y
n
o
m
i
a
l
s
h
a
p
e
c
o
e
f
f
i
c
i
e
n
t
s
a
r
e
e
x
t
r
a
c
t
e
d
f
r
o
m
e
i
t
h
e
r
t
h
e
s
p
e
e
c
h
si
g
n
a
l
,
t
h
e
d
i
sc
r
e
t
e
tr
a
n
s
f
o
r
m
o
f
th
e
s
ig
n
a
l o
r
b
o
th
o
f
th
e
m
.
Bo
t
h
t
h
e
t
r
a
i
n
i
n
g
a
n
d
t
h
e
r
e
c
o
g
n
i
t
i
o
n
m
o
d
e
s
i
n
c
l
u
d
e
f
e
a
t
u
r
e
e
x
t
r
a
c
t
i
o
n
,
s
o
m
e
t
i
m
e
s
c
a
l
l
e
d
t
h
e
f
r
o
n
t
-
en
d
o
f
t
h
e
s
y
s
t
em
.
T
h
e
f
eat
u
r
e
ex
t
r
act
o
r
co
n
v
er
t
s
t
h
e
d
i
g
i
t
al
s
p
eech
s
i
g
n
al
i
n
t
o
a
s
eq
u
en
ce
o
f
num
e
r
i
c
a
l
Evaluation Warning : The document was created with Spire.PDF for Python.
r
IS
S
N
:
2252
-
8938
IJ
-
AI
Vo
l
.
8
, N
o
.
1
,
Ma
r
c
h
20
1
9
:
7
–
13
10
de
s
c
r
i
pt
or
s
,
c
a
l
l
e
d
f
e
a
t
ur
e
ve
c
t
or
s
.
T
he
f
e
a
t
ur
e
s
pr
ovi
de
a
m
or
e
s
t
a
bl
e
,
r
obus
t
,
a
nd
c
om
pa
c
t
r
e
pr
e
s
e
nt
a
t
i
on
th
a
n
th
e
r
a
w
in
p
u
t
s
ig
n
a
l.
F
e
a
tu
r
e
e
x
tr
a
c
tio
n
c
a
n
b
e
c
o
n
s
id
e
r
e
d
a
s
a
d
a
ta
r
e
d
u
c
tio
n
p
r
o
c
e
s
s
th
a
t
a
tte
m
p
ts
to
cap
t
u
r
e
t
h
e
es
s
en
t
i
al
ch
ar
ac
te
r
is
tic
s
o
f
th
e
s
p
e
a
k
e
r
w
ith
a
s
m
a
ll
d
a
ta
r
a
te
[
7
].
D
u
ri
n
g
t
h
e
t
ra
i
n
i
n
g
m
o
d
e
,
e
a
c
h
sp
e
a
k
e
r
i
n
t
h
e
se
t
i
s
m
o
d
e
l
e
d
u
si
n
g
a
se
t
o
f
t
r
a
i
n
i
n
g
d
a
t
a
.
F
e
a
t
u
r
e
s
a
r
e
e
x
t
r
a
c
t
e
d
f
r
o
m
t
h
e
t
r
a
i
n
i
n
g
d
a
t
a
es
s
en
t
i
al
l
y
s
t
r
i
p
i
n
g
aw
ay
al
l
u
n
n
eces
s
ar
y
i
n
f
o
r
m
at
i
o
n
i
n
t
h
e
t
r
ai
n
in
g
s
p
e
e
c
h
s
a
m
p
le
s
le
a
v
in
g
o
n
ly
th
e
s
p
e
a
k
e
r
ch
ar
act
er
i
s
t
i
c
i
n
f
o
r
m
at
i
o
n
,
w
i
t
h
w
h
i
ch
s
p
eak
er
m
o
d
el
s
can
b
e
co
n
s
t
r
u
ct
ed
.
I
n
t
h
e
r
eco
g
n
i
t
i
o
n
m
o
d
e,
f
eat
u
r
es
ar
e
ex
t
r
act
ed
f
r
o
m
t
h
e
u
n
k
n
o
w
n
s
p
eak
er
’
s
v
o
i
ce
s
am
p
l
e.
P
at
t
er
n
m
at
ch
i
n
g
r
ef
er
s
t
o
an
al
g
o
r
i
t
h
m
,
o
r
s
ev
er
al
al
g
o
r
i
t
h
m
s
,
t
h
at
co
m
p
u
t
e
a
m
at
ch
s
co
r
e
b
et
w
een
t
h
e
u
n
k
n
o
w
n
s
p
eak
er
’
s
f
eat
u
r
e
v
ect
o
r
s
an
d
t
h
e
m
o
d
el
s
s
t
o
r
ed
in
th
e
d
a
ta
b
a
s
e
.
T
h
e
o
u
tp
u
t
o
f
th
e
p
a
tte
r
n
m
a
tc
h
in
g
m
o
d
u
le
is
a
s
im
ila
r
ity
s
c
o
r
e
.
T
h
e
la
s
t
p
h
a
s
e
in
th
e
re
c
o
g
n
i
t
i
o
n
c
h
a
i
n
i
s
d
e
c
i
s
i
o
n
m
a
k
i
n
g
.
T
h
e
de
c
i
s
i
on
m
odul
e
t
a
ke
s
t
he
m
a
t
c
h
s
c
or
e
s
a
s
i
t
s
i
nput
,
a
nd
m
a
ke
s
t
he
fi
n
a
l
d
e
c
i
s
i
o
n
o
f
t
h
e
s
p
e
a
k
e
r
i
d
e
n
t
i
t
y
.
It
i
s
c
l
e
a
r
t
h
a
t
t
h
e
fe
a
t
u
re
e
x
t
ra
c
t
i
o
n
p
ro
c
e
s
s
(o
b
t
a
i
n
i
n
g
s
p
e
a
k
e
r
di
s
c
r
i
m
i
na
t
or
y
i
nf
or
m
a
t
i
on)
a
nd
t
he
c
l
a
s
s
i
f
i
c
a
t
i
on
pr
oc
e
s
s
(
us
i
ng
t
he
f
e
a
t
ur
e
s
t
o
de
t
e
r
m
i
ne
t
he
c
or
r
e
c
t
sp
e
a
k
e
r
)
a
l
g
o
r
i
t
h
m
s
a
r
e
o
f
c
r
i
t
i
c
a
l
i
m
p
o
r
t
a
n
c
e
t
o
a
n
y
sp
e
a
k
e
r
i
d
e
n
t
i
f
i
c
a
t
i
o
n
sy
st
e
m
.
4.
1.
Fe
a
t
u
r
e
e
x
t
r
a
c
t
i
o
n
A
s
i
n
g
l
e
h
u
m
a
n
s
p
e
e
c
h
s
i
g
n
a
l
c
o
n
t
a
i
n
s
a
l
a
r
g
e
a
m
o
u
n
t
o
f
s
p
e
a
k
e
r
d
e
p
e
n
d
e
n
t
i
n
f
o
r
m
a
t
i
o
n
.
W
h
i
l
e
t
h
e
hum
a
n
br
a
i
n
i
s
a
bl
e
t
o
di
s
tin
g
u
is
h
b
e
tw
e
e
n
s
p
e
a
k
e
r
s
b
a
s
e
d
o
n
‘
h
ig
h
-
le
v
e
l’
p
r
o
p
e
r
tie
s
s
u
c
h
a
s
d
ia
le
c
t, s
p
e
a
k
in
g
st
y
l
e
,
c
o
n
t
e
x
t
o
f
t
h
e
sp
e
e
c
h
a
n
d
t
h
e
e
m
o
t
i
o
n
a
l
st
a
t
e
o
f
t
h
e
sp
e
a
k
e
r
,
d
e
si
g
n
i
n
g
i
d
e
n
t
i
f
i
c
a
t
i
o
n
a
l
g
o
r
i
t
h
m
s
b
a
se
d
on
t
he
s
e
pr
ope
r
t
i
e
s
i
s
i
nf
e
a
s
i
bl
e
due
t
o
t
he
r
e
qui
r
e
d
hi
gh
c
om
pl
e
xi
t
y.
I
t
i
s
pos
s
i
bl
e
how
e
ve
r
t
o
bui
l
d
e
f
f
i
c
i
e
nt
id
e
n
tif
ic
a
tio
n
a
lg
o
r
ith
m
s
b
a
s
e
d
o
n
th
e
lo
w
-
le
v
e
l
p
r
o
p
e
r
tie
s
o
f
th
e
s
ig
n
a
l
s
u
c
h
a
s
p
itc
h
,
in
te
n
s
ity
,
f
o
r
m
a
n
t
fre
q
u
e
n
c
i
e
s
a
n
d
t
h
e
i
r
c
h
a
ra
c
t
e
ri
s
t
i
c
s
.
T
h
e
c
o
n
c
e
p
t
o
f
fe
a
t
u
re
e
x
t
ra
c
t
i
o
n
c
o
n
t
ri
b
u
t
e
s
t
o
t
h
e
g
o
a
l
o
f
id
e
n
tif
y
in
g
sp
e
a
k
e
r
s
b
a
se
d
o
n
t
h
e
l
o
w
-
le
v
e
l
p
r
o
p
e
r
tie
s
in
tw
o
w
a
y
s
.
F
ir
s
tly
,
th
e
e
x
tr
a
c
tio
n
p
r
o
d
u
c
e
s
s
u
f
f
ic
ie
n
t
in
f
o
r
m
a
tio
n
f
o
r
g
o
o
d
s
p
e
a
k
e
r
d
is
c
r
im
in
a
tio
n
a
n
d
c
a
p
tu
r
e
s
th
is
in
f
o
r
m
a
tio
n
in
a
f
o
r
m
a
n
d
s
iz
e
th
a
t
a
llo
w
ef
f
i
ci
en
t
m
o
d
el
i
n
g
.
Se
c
o
n
d
l
y
,
f
e
a
t
u
re
e
x
t
ra
c
t
i
o
n
c
a
n
b
e
c
o
n
s
i
d
e
re
d
a
s
a
d
a
t
a
re
d
u
c
t
i
o
n
p
ro
c
e
s
s
t
h
a
t
a
t
t
e
m
p
t
s
t
o
cap
t
u
r
e
t
h
e
es
s
en
t
i
al
ch
ar
act
er
i
s
t
i
cs
o
f
t
h
e
s
p
eak
er
w
i
t
h
a
s
m
al
l
d
at
a
r
at
e.
T
h
e
f
eat
u
r
e
ex
t
r
act
o
r
co
n
v
er
t
s
t
h
e
di
gi
t
a
l
s
pe
e
c
h
s
i
gna
l
i
nt
o
a
s
e
que
nc
e
of
num
e
r
i
c
a
l
de
s
c
r
i
pt
or
s
c
a
l
l
ed
f
eat
u
r
e
v
ect
o
r
s
.
S
ev
er
al
f
eat
u
r
e
ex
t
r
act
i
o
n
t
ech
n
i
q
u
es
ar
e
u
s
ed
i
n
s
p
eak
er
r
eco
g
n
i
t
i
o
n
s
y
s
t
em
s
.
T
h
e
co
n
cep
t
o
f
f
eat
u
r
e
ex
t
r
act
i
o
n
u
s
i
n
g
t
h
e
MF
C
C
s
i
s
w
i
d
e
l
y
k
n
o
w
n
i
n
s
p
e
a
k
e
r
i
d
e
n
t
i
f
i
c
a
t
i
o
n
,
I
t
c
o
n
t
r
i
b
u
t
e
s
t
o
t
h
e
g
o
a
l
o
f
i
d
e
n
t
i
f
y
i
n
g
s
p
e
a
k
e
r
s
b
a
s
e
d
o
n
th
e
lo
w
-
le
v
e
l
p
r
o
p
e
r
tie
s
[
8
].
It
i
s
c
l
e
a
r
t
h
a
t
t
h
e
s
p
e
e
c
h
s
i
g
n
a
l
h
a
s
o
s
c
i
l
l
a
t
o
ry
p
a
t
t
e
rn
s
,
w
h
i
c
h
s
u
p
p
o
rt
s
t
h
e
ap
p
l
i
cat
i
o
n
o
f
t
h
e
cep
s
t
r
al
m
et
h
o
d
f
o
r
f
eat
u
r
e
ex
t
r
act
i
o
n
f
r
o
m
o
u
r
s
p
eech
s
i
g
n
al
s
.
I
n
s
p
eak
er
i
d
en
t
i
f
i
cat
i
o
n
,
th
e
e
x
tr
a
c
tio
n
p
r
o
d
u
c
e
s
s
u
f
f
ic
ie
n
t
in
f
o
r
ma
t
i
o
n
f
o
r
g
o
o
d
s
p
e
a
k
e
r
d
i
s
c
r
i
mi
n
a
t
i
o
n
.
I
n
t
h
e
f
o
l
l
o
w
i
n
g
s
u
b
s
e
c
t
i
o
n
,
an
ex
p
l
an
at
i
o
n
f
o
r
t
h
e
ex
t
r
act
i
o
n
o
f
t
h
e
M
F
C
C
s
an
d
t
h
e
p
o
l
y
n
o
m
i
al
co
ef
f
i
ci
en
t
s
i
s
p
r
es
en
t
ed
.
4.
2.
Ex
t
r
a
c
t
i
o
n
o
f
M
F
C
C
s
Th
e
M
F
C
C
s
a
r
e
c
o
m
m
o
n
l
y
e
x
t
r
a
c
t
e
d
f
r
o
m
s
p
e
e
c
h
s
i
g
n
a
l
s
t
h
r
o
u
g
h
c
e
p
s
t
ra
l
a
n
a
l
y
s
i
s
.
T
h
e
i
n
p
u
t
s
i
g
n
a
l
is
f
ir
s
t f
r
a
m
e
d
a
n
d
w
in
d
o
w
e
d
, th
e
F
o
u
r
ie
r
tr
a
n
s
f
o
r
m
is
th
e
n
ta
k
e
n
a
n
d
th
e
m
a
g
n
itu
d
e
o
f
th
e
r
e
s
u
ltin
g
s
p
e
c
tr
u
m
is
w
a
r
p
e
d
b
y
th
e
M
e
l
-
sc
a
l
e
.
T
h
e
l
o
g
o
f
t
h
i
s
sp
e
c
t
r
u
m
i
s
t
h
e
n
t
a
k
e
n
a
n
d
t
h
e
D
C
T
i
s
a
p
p
l
i
e
d
sh
o
w
n
in
Fi
g
u
r
e
5
.
Th
e
1
-
D
s
i
g
n
a
l
m
u
s
t
f
i
r
s
t
b
e
b
r
o
k
e
n
u
p
i
n
t
o
s
m
a
l
l
s
e
c
t
i
o
n
s
;
e
a
c
h
o
f
N
s
a
m
p
l
e
s
.
T
h
e
s
e
s
e
c
t
i
o
n
s
a
r
e
cal
l
ed
f
r
am
es
an
d
t
h
e
m
o
t
i
v
at
i
o
n
f
o
r
t
h
i
s
f
r
am
i
n
g
p
r
o
ces
s
i
s
t
h
e
q
u
as
i
s
t
at
i
o
n
ar
y
n
at
u
r
e
o
f
t
h
e
1
-
D
s
i
g
n
a
l
s
.
Ho
we
v
e
r
,
i
f
we
e
x
a
m
i
n
e
t
h
e
s
i
g
n
a
l
o
v
e
r
d
i
s
c
r
e
t
e
s
e
c
t
i
o
n
s
,
w
h
ic
h
a
r
e
s
u
f
f
ic
ie
n
tly
s
h
o
r
t
in
d
u
r
a
tio
n
,
th
e
n
th
e
s
e
se
c
t
i
o
n
s
c
a
n
b
e
c
o
n
si
d
e
r
e
d
a
s
st
a
t
i
o
n
a
r
y
a
n
d
e
x
h
i
b
i
t
st
a
b
l
e
c
h
a
r
a
c
t
e
r
i
st
i
c
s
[9
-
11]
.
To
a
v
o
i
d
l
o
s
s
o
f
in
f
o
r
m
a
tio
n
,
f
r
a
m
e
o
v
e
r
la
p
is
u
s
e
d
.
E
a
c
h
f
r
a
m
e
b
e
g
in
s
a
t
s
o
m
e
o
f
f
s
e
t
o
f
L
s
a
m
p
le
s
w
ith
r
e
s
p
e
c
t
to
t
he
pr
e
vi
ous
f
r
a
m
e
w
he
r
e
L
B
N
.
F
or
e
a
c
h
f
r
a
m
e
,
a
w
i
ndow
i
ng
f
unc
t
i
on
i
s
us
ua
l
l
y
a
ppl
i
e
d
t
o
i
nc
r
e
a
s
e
t
he
co
n
t
i
n
u
i
t
y
b
et
w
een
ad
j
acen
t
f
r
am
es
.
C
o
m
m
o
n
w
i
n
d
o
w
i
n
g
f
u
n
ct
i
o
n
s
i
n
cl
u
d
e
t
h
e
r
ect
an
g
u
l
ar
w
i
n
d
o
w
,
t
h
e
Ha
m
m
i
n
g
wi
n
d
o
w,
t
h
e
B
l
a
c
k
m
a
n
wi
n
d
o
w
a
n
d
f
l
a
t
t
o
p
wi
n
d
o
w.
Wi
n
d
o
w
i
n
g
i
n
t
i
m
e
d
o
m
a
i
n
i
s
a
p
o
i
n
t
w
i
s
e
mu
l
t
i
p
l
i
c
a
t
i
o
n
o
f
t
h
e
f
r
a
me
a
n
d
t
h
e
w
i
n
d
o
w
f
u
n
c
t
i
o
n
.
T
h
e
ma
g
n
i
t
u
d
e
s
p
e
c
t
r
u
m
X
k
is
n
o
w
s
c
a
le
d
in
b
o
th
fre
q
u
e
n
c
y
a
n
d
m
a
g
n
i
t
u
d
e
.
F
i
rs
t
,
t
h
e
fre
q
u
e
n
c
y
i
s
s
c
a
l
e
d
l
o
g
a
ri
t
h
m
i
c
a
l
l
y
u
s
i
n
g
t
h
e
s
o
-
cal
l
ed
M
el
f
i
l
t
er
b
an
k
H
k
,
m
, a
nd
t
he
n
t
he
l
oga
r
i
t
hm
i
s
t
a
ke
n,
gi
vi
ng:
′
=
ln
k
.
H
,
fo
r
m
=
1
,
2
,
…
,
M
,
w
h
e
re
M
i
s
t
h
e
n
u
m
b
e
r
o
f
fi
l
t
e
r
b
a
n
k
s
a
n
d
M
˂
˂
N
.
Th
e
M
e
l
f
i
l
t
e
r
b
a
n
k
i
s
a
c
o
l
l
e
c
t
i
o
n
o
f
t
r
i
a
n
g
u
l
a
r
f
i
l
t
e
r
s
d
e
f
i
n
e
d
b
y
c
e
n
t
e
r
f
r
e
q
u
e
n
c
i
e
s
c
a
l
c
u
l
a
t
e
d
o
n
th
e
M
e
l
s
c
a
le
(
S
r
in
iv
a
s
a
n
e
t
a
l.
2
0
0
4
;
L
u
n
g
y
u
n
e
t
a
l.
2
0
0
6
)
.
T
h
e
tr
ia
n
g
u
la
r
f
ilte
r
s
a
r
e
s
p
r
e
a
d
o
v
e
r
th
e
e
n
tir
e
fre
q
u
e
n
c
y
ra
n
g
e
fro
m
z
e
ro
t
o
t
h
e
N
y
q
u
i
s
t
fr
eq
u
en
cy
.
T
h
e
n
u
m
b
er
o
f
f
i
l
t
er
s
i
s
o
n
e
o
f
t
h
e
p
ar
am
et
er
s
w
h
i
ch
Evaluation Warning : The document was created with Spire.PDF for Python.
IJ
-
AI
IS
S
N
:
2252
-
8938
r
Spe
e
c
h
r
e
c
ogni
t
i
on
of
m
or
oc
c
an
di
al
e
c
t
us
i
ng
hi
dde
n
ma
r
k
o
v
mo
d
e
l
s
(
Be
z
o
u
i
M
o
u
a
z
)
11
af
f
ect
t
h
e
r
eco
g
n
i
t
i
o
n
accu
r
acy
o
f
t
h
e
s
y
s
t
em
.
F
i
n
al
l
y
,
t
h
e
M
F
C
C
s
ar
e
o
b
t
ai
n
ed
b
y
co
m
p
u
t
i
n
g
t
h
e
D
C
T
o
f
X′
m
us
i
ng:
=
′
co
s
−
1
2
fo
r
l
=
1
,
2
,
…
,
M
,
w
h
e
re
is
th
e
lth
MF
C
C
.
F
ig
u
r
e
5.
Th
e
M
e
l
f
i
l
t
e
r
b
a
n
k
Th
e
n
u
m
b
e
r
o
f
t
h
e
re
s
u
l
t
i
n
g
M
F
C
C
s
i
s
c
h
o
s
e
n
b
e
t
w
e
e
n
1
2
a
n
d
2
0
,
s
i
n
c
e
m
o
s
t
o
f
t
h
e
s
i
g
n
a
l
in
f
o
r
m
a
tio
n
is
r
e
p
r
e
s
e
n
te
d
b
y
th
e
f
ir
s
t
f
e
w
c
o
e
f
f
ic
ie
n
ts
.
T
h
e
0
th
c
o
e
f
f
ic
ie
n
t
r
e
p
r
e
s
e
n
ts
th
e
a
v
e
r
a
g
e
lo
g
e
n
e
r
g
y
of
t
he
f
r
a
m
e
.
A
n
e
xpe
r
i
e
nc
e
w
i
t
h
s
pe
e
c
h
r
e
c
ogni
t
i
on[
12
-
13]
,
sh
o
w
e
d
t
h
a
t
i
t
is
b
e
n
e
f
ic
ia
l
to
u
s
e
a
ls
o
d
e
lta
a
n
d
de
l
t
a
-
de
l
t
a
c
oe
f
f
i
c
i
e
nt
s
w
hi
c
h
de
c
r
e
a
s
e
t
he
w
or
d
e
r
r
or
r
a
t
e
.
E
ve
n
t
hough
t
he
or
i
gi
na
l
s
e
t
of
f
e
a
t
ur
e
s
of
t
he
MF
C
C
i
s
m
o
r
e
o
r
l
e
s
s
c
o
r
r
e
l
a
t
e
d
t
h
e
n
a
f
t
e
r
a
d
d
i
t
i
o
n
o
f
d
e
l
t
a
a
n
d
d
e
l
t
a
-
de
l
t
a
f
e
a
t
ur
e
s
t
he
i
nf
or
m
a
t
i
on
re
d
u
n
d
a
n
cy
o
f
el
em
en
t
s
i
n
f
eat
u
r
e
v
ect
o
r
s
i
n
cr
eas
es
.
S
i
n
ce
i
n
t
h
i
s
s
y
s
t
em
w
e
ar
e
co
n
cer
n
ed
w
i
t
h
t
h
e
s
p
ect
r
al
fe
a
t
u
re
s
s
u
c
h
a
s
M
F
C
C
fe
a
t
u
re
s
,
w
e
a
d
d
d
i
ffe
re
n
t
fe
a
t
u
re
s
re
l
a
t
e
d
t
o
t
h
e
M
F
C
C
s
u
c
h
a
s
t
i
m
e
d
e
ri
v
a
t
i
v
e
s
[
14
].
Th
e
f
i
r
s
t
o
r
d
e
r
r
e
g
r
e
s
s
i
o
n
c
o
e
f
f
i
c
i
e
n
t
s
o
f
t
h
e
MF
C
C
f
e
a
t
u
r
e
v
e
c
t
o
r
c
a
l
l
e
d
D
e
l
t
a
i
s
i
n
c
l
u
d
e
d
.
A
l
s
o
,
t
h
e
s
e
c
o
n
d
or
de
r
r
e
gr
e
s
s
i
on
c
oe
f
f
i
c
i
e
nt
s
,
c
a
l
l
e
d
D
e
l
t
a
-
De
l
t
a
,
i
s
i
n
c
l
u
d
e
d
.
5.
RE
S
UL
T
S
AND
DI
S
CUS
S
I
O
N
Th
e
d
a
t
a
b
a
s
e
u
s
e
d
i
n
t
r
a
i
n
i
n
g
a
n
d
t
e
s
t
i
n
g
t
h
e
s
y
s
t
e
m
f
o
r
e
a
c
h
d
i
a
l
e
c
t
i
s
a
c
o
m
b
i
n
a
t
i
o
n
o
f
t
w
e
n
t
y
sp
e
a
k
e
r
s,
e
l
e
v
e
n
m
a
l
e
s
a
n
d
n
i
n
e
f
e
m
a
l
e
s.
W
e
c
h
o
se
4
p
e
o
p
l
e
,
3
w
o
m
e
n
a
n
d
a
2
m
e
n
s
t
o
p
r
o
n
o
u
n
c
e
4
w
o
r
d
s
of
t
he
M
or
oc
c
a
n
di
a
l
e
c
t
w
hi
c
h
a
r
e
:
(
مﻼ
ﺳ
)
=
(H
i
),
(
ﺮﯾ
ا
ﺪ
ﯿ
ﻛ
)
=
(H
o
w
a
re
y
o
u
),
(
سﺎ
ﺑ
ﻻ
)
=
(T
h
e
re
i
s
n
o
t
h
i
n
g
w
ro
n
g
),
(
ﺮﯿ
ﺨ
ﺑ
)
=
(F
i
n
e
)
a
n
d
w
e
re
c
o
rd
e
d
t
h
e
v
o
i
c
e
s
of
t
he
s
pe
a
ke
r
s
i
n
f
i
l
e
s
i
n
(
.
w
a
v)
f
or
m
a
t
.
l
a
t
e
r
w
e
be
ga
n
our
w
or
k
of
t
he
r
e
c
ogni
t
i
on
a
nd
t
he
l
yr
i
c
s
of
t
he
M
or
oc
c
a
n
di
a
l
e
c
t
by
t
he
pa
r
t
L
e
a
r
ni
ng
[
11]
,
t
hr
ough
'
'
a
dd
a
ne
w
s
ound
fro
m
fi
l
e
'
'
w
h
i
c
h
i
n
v
i
t
e
s
t
h
e
u
s
e
r
t
o
c
h
o
o
s
e
a
fi
l
e
(.
w
a
v
)
a
n
d
c
l
a
s
s
i
fy
i
t
b
y
Id
e
n
t
i
t
y
,
fro
m
ID
:
1
t
o
ID
:
5
[1
5
].
w
e
co
n
t
i
n
u
ed
t
h
e
t
r
ai
n
i
n
g
p
h
as
e
t
o
b
u
i
l
d
a
d
at
ab
as
e
o
f
f
i
l
es
(
.
w
av
)
w
i
t
h
4
cl
as
s
es
,
each
cl
as
s
r
ep
r
es
en
t
s
a
w
el
l
-
de
f
i
ne
d
s
pe
a
ke
r
.
T
he
s
pe
e
c
h
of
e
l
e
ve
n
m
a
l
e
s
pe
a
ke
r
s
a
nd
ni
ne
f
e
m
a
l
e
s
pe
a
ke
r
s
a
r
e
us
e
d
f
or
t
r
a
i
ni
ng,
a
nd
t
he
s
pe
e
c
h
of
one
m
a
l
e
s
pe
a
ke
r
a
nd
t
hr
e
e
f
e
m
a
l
e
s
pe
a
ke
r
s
a
r
e
us
e
d
f
or
t
e
s
t
i
ng.
T
he
s
pe
e
c
h
f
or
t
r
a
i
ni
ng
f
r
om
e
a
c
h
sp
e
a
k
e
r
i
s
o
n
e
m
i
n
u
t
e
l
o
n
g
.
T
h
e
sp
e
e
c
h
f
o
r
t
e
st
i
n
g
f
r
o
m
e
a
c
h
sp
e
a
k
e
r
i
s
1
0
se
c
o
n
d
l
o
n
g
[1
6
].
O
u
r
H
M
M
Sp
e
e
c
h
R
e
c
o
g
n
i
t
i
o
n
Sy
s
t
e
m
w
a
s
p
r
o
g
r
a
m
m
e
d
b
y
u
s
i
ng
M
A
T
L
A
B
t
o
c
r
e
a
t
e
a
us
e
r
i
nt
e
r
f
a
c
e
a
nd
t
o
e
na
bl
e
us
e
r
t
o
a
dd
a
ne
w
s
ound
f
r
om
au
d
i
o
f
i
l
es
as
s
h
o
w
n
i
n
Fi
g
u
r
e
6
.
Fi
g
u
r
e
6.
Fl
o
w
c
h
a
r
t
o
f
t
h
e
d
e
e
p
l
e
a
r
n
i
n
g
m
o
d
e
l
.
Evaluation Warning : The document was created with Spire.PDF for Python.
r
IS
S
N
:
2252
-
8938
IJ
-
AI
Vo
l
.
8
, N
o
.
1
,
Ma
r
c
h
20
1
9
:
7
–
13
12
W
e
p
r
es
en
t
i
n
th
is
s
e
c
tio
n
th
e
r
e
s
u
lts
o
f
th
e
id
e
n
tif
ic
a
tio
n
p
r
o
c
e
s
s
in
c
lu
d
in
g
th
e
te
s
tin
g
r
e
s
u
lts
.
I
t
s
h
o
u
ld
b
e
me
n
t
i
o
n
e
d
t
h
a
t
n
o
d
e
t
a
i
l
e
d
o
r
w
o
r
d
-
le
v
e
l
la
b
e
llin
g
w
a
s
d
o
n
e
f
o
r
th
e
d
a
ta
b
a
s
e
r
e
la
te
d
to
th
e
te
s
tin
g
s
te
p
[
1
7
];
sp
e
e
c
h
i
s
l
a
b
e
l
l
e
d
[1
8
]
a
c
c
o
rd
i
n
g
t
o
t
h
e
c
o
rre
sp
o
n
d
i
n
g
d
i
a
l
e
c
t
d
a
t
a
.
F
o
r
i
n
st
a
n
c
e
,
a
M
o
r
o
c
c
a
n
sp
e
e
c
h
f
i
l
e
i
s
la
b
e
lle
d
w
ith
th
e
le
tte
r
"
M
"
w
h
ile
a
M
S
A
s
p
e
e
c
h
f
ile
is
la
b
e
lle
d
w
ith
th
e
le
tte
r
"
A
"
.
T
r
a
in
in
g
w
ith
b
o
th
M
S
A
an
d
M
o
r
o
ccan
D
i
al
ect
as
s
h
o
w
n
i
n
Ta
b
l
e
1
.
We
f
o
l
l
o
w
e
d
t
h
r
e
e
w
a
y
s
o
f
t
r
e
a
t
m
e
n
t
f
o
r
t
r
ai
n
i
n
g
s
t
ep
:
-
Tr
a
i
n
i
n
g
w
i
t
h
M
S
A
f
i
l
e
s
(
s
p
e
a
k
e
r
s
)
:
Th
e
s
y
s
t
e
m
c
o
u
l
d
i
d
e
n
t
i
f
y
a
l
l
M
S
A
t
e
s
t
i
n
g
f
i
l
e
s
a
n
d
n
o
t
h
i
n
g
o
f
MO
R
O
C
C
A
N
f
i
l
e
s
.
-
Tr
a
i
n
i
n
g
w
i
t
h
M
o
r
o
c
c
a
n
f
i
l
e
s
:
Th
e
s
y
s
t
e
m
c
o
u
l
d
i
d
e
n
t
i
f
y
a
l
l
M
O
R
O
C
C
A
N
t
e
s
t
i
n
g
f
i
l
e
s
a
n
d
n
o
t
h
i
n
g
fro
m
M
S
A
fi
l
e
s
.
-
Tr
a
i
n
i
n
g
w
i
t
h
b
o
t
h
M
S
A
a
nd
M
or
oc
c
a
n
f
i
l
e
s
:
T
he
s
ys
t
e
m
c
oul
d
i
de
nt
i
f
y
a
l
l
10
t
e
s
t
i
ng
f
i
l
e
s
of
M
S
A
a
nd
8
f
i
l
e
s
f
r
om
10
t
e
s
t
i
ng
f
i
l
e
s
of
M
or
oc
c
a
n
D
i
a
l
e
c
t
.
Ta
b
l
e
1
.
Tr
a
i
n
i
n
g
w
i
t
h
b
o
t
h
M
S
A
a
n
d
M
o
r
o
c
c
a
n
D
i
a
l
e
c
t
Sp
e
a
k
e
r
s
Nu
mb
e
r
o
f
a
t
t
e
mp
t
s
Re
c
o
g
n
i
z
e
d
w
o
r
d
Sa
l
a
m
مﻼ
ﺳ
Ki
d
a
y
r
ﺮﯾ
ا
ﺪ
ﯿ
ﻛ
La
b
a
s
سﺎ
ﺑ
ﻻ
Bi
k
h
i
r
ﺮﯿ
ﺨﺑ
ID
1
4
3
ID
2
4
4
ID
3
ID
4
ID
5
To
t
a
l
4
4
4
20
4
3
4
18
6.
C
ON
C
L
U
S
I
ON
Th
e
p
u
r
p
o
s
e
o
f
t
h
i
s
w
o
r
k
i
s
t
o
v
e
r
i
f
y
t
h
e
a
b
i
l
i
t
y
o
f
o
u
r
H
M
M
S
p
e
e
c
h
R
e
c
o
g
n
i
t
i
o
n
S
y
s
t
e
m
t
o
di
s
t
i
ngui
s
h
t
he
voc
a
l
pr
i
nt
of
s
pe
a
ke
r
s
,
a
nd
i
de
nt
i
f
y
t
he
m
by
gi
vi
ng
e
a
c
h
of
t
he
m
a
s
pe
c
i
f
i
c
c
l
a
s
s
.
T
hi
s
i
s
done
th
r
o
u
g
h
c
r
e
a
te
a
s
p
e
e
c
h
r
e
c
o
g
n
itio
n
s
y
s
te
m
,
a
n
d
a
p
p
ly
it
to
a
M
o
r
o
c
c
a
n
D
ia
le
c
t
s
p
e
e
c
h
[1
9
-
20
].
B
y
in
v
e
s
tig
a
tin
g
th
e
e
x
tr
a
c
te
d
f
e
a
tu
r
e
s
o
f
th
e
u
n
k
n
o
w
n
s
p
e
e
c
h
a
n
d
th
e
n
c
o
m
p
a
r
e
th
e
m
to
th
e
s
to
r
e
d
e
x
tr
a
c
te
d
fe
a
t
u
re
v
e
c
t
o
rs
fo
r
e
a
c
h
d
i
ffe
re
n
t
s
p
e
a
k
e
r,
i
n
o
rd
e
r,
t
o
i
d
e
n
t
i
fy
t
h
e
u
n
k
n
o
w
n
s
p
e
a
k
e
r.
T
h
e
m
o
d
e
l
u
s
ed
i
n
t
h
i
s
wo
r
k
wa
s
t
h
e
Hi
d
d
e
n
M
a
r
k
o
v
M
o
d
e
l
[
21
].
Th
e
M
F
C
C
+
D
e
l
t
a
+
D
e
l
t
a
-
De
l
t
a
f
e
a
t
u
r
e
s
p
e
r
f
o
r
m
e
d
b
e
s
t
re
a
c
h
i
n
g
a
n
i
d
e
n
t
i
fi
c
a
t
i
o
n
s
c
o
re
.
T
h
e
a
c
c
u
ra
c
y
o
f
our
HM
M
S
R
S
(H
M
M
S
p
e
e
c
h
R
e
c
o
g
n
i
t
i
o
n
S
y
s
t
e
m
)
is
a
b
o
u
t
90%
.
ACK
NO
W
L
E
DG
E
M
E
NT
S
We
w
o
u
l
d
l
i
k
e
t
o
a
c
k
n
o
w
l
e
d
g
e
t
h
e
m
a
i
n
s
i
t
e
w
h
e
r
e
r
e
s
e
a
r
c
h
w
a
s
c
a
r
r
i
e
d
o
u
t
;
D
e
p
a
r
t
m
e
n
t
o
f
Co
m
p
u
t
e
r
,
Ch
o
u
a
i
b
D
o
u
k
k
a
l
i
U
n
i
v
e
r
s
i
t
y
,
F
a
c
u
l
t
y
o
f
S
c
i
e
n
c
e
,
E
I
J
a
d
i
d
a
,
M
o
r
o
c
c
o
.
RE
F
E
RE
NCE
S
[1
]
Mo
h
a
m
e
d
A
m
i
n
e
Me
n
a
c
e
r
,
O
d
i
l
e
Me
l
l
a
,
D
o
m
i
n
i
q
u
e
F
o
h
r
,
D
e
n
i
s
J
o
u
v
e
t,
D
a
v
id
L
a
n
g
lo
is
,
K
a
m
e
l
S
m
a
ili.
"D
ev
el
o
p
m
en
t
o
f
t
h
e
A
r
ab
i
c
L
o
r
i
a
A
u
t
o
m
at
i
c
S
p
eech
R
eco
g
n
i
t
i
o
n
sy
st
em
(
A
L
A
S
R
)
an
d
i
t
s
ev
al
u
at
i
o
n
f
o
r
A
l
g
er
i
an
di
a
l
e
c
t
,
"
3r
d
I
nt
e
r
na
t
i
ona
l
C
onf
e
r
e
nc
e
on
A
r
a
bi
c
C
om
put
a
t
i
ona
l
L
i
ngui
s
t
i
c
s
,
A
C
L
i
ng
2017,
5
–
6
N
ove
m
be
r
2017,
Du
b
a
i
,
Un
i
t
e
d
Ar
a
b
E
mi
r
a
t
e
s
.
[2
]
S.
C
o
n
n
e
l
,
A
C
o
m
p
a
r
i
s
o
n
o
f
H
i
d
d
e
n
M
a
r
k
o
v
M
o
d
e
l
Fe
a
t
u
r
e
s
f
o
r
t
h
e
R
e
c
o
g
n
i
t
i
o
n
o
f
C
u
r
s
i
v
e
H
a
n
d
w
r
i
t
i
n
g
,
Co
m
p
u
t
e
r
S
c
i
e
n
c
e
D
e
p
a
r
t
m
e
n
t
,
M
i
c
h
i
g
a
n
S
t
a
t
e
U
n
i
v
e
r
s
i
t
y
,
M
S
T
h
e
s
i
s
(
1
9
96)
.
[3
]
Ra
b
i
n
e
r
L
-
R.
,
J
u
a
n
g
B
-
H.
,
F
u
n
d
a
me
n
t
a
l
s
o
f
S
p
e
e
c
h
R
e
c
o
g
n
i
t
i
o
n
,
P
r
e
n
t
i
c
e
-
Ha
l
l
,
1
9
9
3
.
[4
]
Gr
a
v
e
s
A,
M
o
h
a
me
d
A,
Hi
n
t
o
n
G
(
2
0
1
3
)
"
S
p
e
e
c
h
r
e
c
o
g
n
i
t
i
o
n
wi
t
h
d
e
e
p
r
e
c
u
r
r
e
n
t
n
e
u
r
a
l
n
e
t
wo
r
k
s
"
,
I
n
I
C
AS
S
P
2013,
pp.
1
-
5.
[5
]
Sr
i
n
i
v
a
s
a
n
e
t
Th
o
r
p
e
a
n
d
Sh
e
l
t
o
n
1
9
9
3
e
t
a
l
.
2
0
0
4
;
L
u
n
g
y
u
n
e
t
a
l.
2
0
0
6
;
C
h
iy
i
a
n
d
K
ubi
c
he
k
1996;
L
a
m
e
t
a
l
.
1996.
[6
]
Af
i
f
y
,
M
.
,
S
a
r
i
k
a
y
a
,
R
.
,
k
wa
n
g
J
e
f
f
Ku
o
,
H.
,
B
e
s
a
c
i
e
r
,
L
.
,
Ga
o
,
Y.
,
2
0
0
6
.
"
On
t
h
e
u
s
e
o
f
mo
r
p
h
o
l
o
g
i
c
a
l
a
n
a
l
y
s
i
s
f
o
r
di
a
l
e
c
t
a
l
a
r
a
bi
c
s
pe
e
c
h
r
e
c
ogni
t
i
on"
,
i
n:
P
r
oc
e
ed
i
n
g
s
o
f
I
C
S
L
P
0
6
,
p
p
.
277
-
280.
[7
]
Ad
n
a
n
Qa
y
y
u
m,
S
i
d
d
i
q
u
e
L
a
t
i
f
,
J
u
n
a
i
d
Qa
d
i
r
,
"
Qu
r
a
n
R
e
c
i
t
e
r
I
d
e
n
t
i
f
i
c
a
t
i
o
n
:
A
De
e
p
L
e
a
r
n
i
n
g
Ap
p
r
o
a
c
h
"
,
Co
m
p
u
t
e
r
a
n
d
Co
m
m
u
n
i
c
a
t
i
o
n
E
n
g
i
n
e
e
r
i
n
g
(
I
CCCE
)
2
0
1
8
7
t
h
I
n
t
e
r
n
a
t
i
o
n
a
l
Co
n
f
e
r
e
n
c
e
o
n
,
p
p
.
4
9
2
-
497,
2018.
[8
]
Sa
l
i
m
a
H
a
r
r
a
t
,
K
a
r
i
m
a
M
e
f
t
o
u
h
,
M
o
u
r
a
d
A
b
ba
s
,
K
ha
l
e
d
-
Wa
l
i
d
H
i
d
o
u
c
i
a
n
d
K
a
m
e
l
S
m
a
i
l
i
,
"
A
n
A
l
g
e
r
i
a
n
d
i
a
l
e
c
t
:
St
u
d
y
a
n
d
R
e
s
o
u
r
c
e
s
"
,
Pu
b
l
i
s
h
e
d
i
n
:
I
J
A
C
SA
(
I
n
t
e
r
n
a
t
i
o
n
a
l
J
o
u
r
n
a
l
o
f
A
d
v
a
n
c
e
d
C
o
m
p
u
t
e
r
Sc
i
e
n
c
e
a
n
d
Ap
p
l
i
c
a
t
i
o
n
s
)
,
Vo
l
.
7
,
No
.
3
,
2
0
1
6
.
Evaluation Warning : The document was created with Spire.PDF for Python.
IJ
-
AI
IS
S
N
:
2252
-
8938
r
Spe
e
c
h
r
e
c
ogni
t
i
on
of
m
or
oc
c
an
di
al
e
c
t
us
i
ng
hi
dde
n
ma
r
k
o
v
mo
d
e
l
s
(
Be
z
o
u
i
M
o
u
a
z
)
13
[9
]
Ta
c
h
i
c
a
r
t
R
i
d
o
u
a
n
e
a
n
d
B
o
u
z
o
u
b
a
a
K
a
r
i
m
.
2
0
1
4
.
"
A
h
y
b
r
i
d
ap
p
r
o
ach
t
o
t
r
an
sl
at
e
m
o
r
o
ccan
ar
ab
i
c
d
i
al
ect
".
I
n
SI
TA
’
1
4
,
9
t
h
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
I
n
t
e
l
l
i
g
e
n
t
Sy
s
t
e
m
s
.
[1
0
]
Mo
u
a
z
B
e
z
o
u
i
,
A
b
d
e
l
m
a
j
i
d
E
l
m
o
u
t
a
o
u
a
k
k
i
l
,
a
n
d
A
b
d
e
r
r
a
h
i
m
B
e
n
i
h
s
s
a
n
e
.
F
e
a
t
u
r
e
e
x
t
r
a
c
t
i
o
n
o
f
s
o
m
e
Q
u
r
a
n
i
c
re
c
i
t
a
t
i
o
n
u
s
i
n
g
m
e
l
fre
q
u
e
n
c
y
c
e
p
s
t
ra
l
co
ef
i
ci
en
t
s
(
M
F
C
C
)
.
I
n
M
u
l
t
i
m
ed
i
a
C
o
m
p
u
t
i
n
g
an
d
S
y
st
em
s
(
I
C
M
C
S
)
,
2
0
1
6
5t
h
I
nt
e
r
na
t
i
ona
l
C
onf
e
r
e
nc
e
on,
pa
ge
s
127
–
131.
I
E
E
E
,
2016.
[1
1
]
Jy
o
t
i
G
u
g
l
an
i
,
A
.
N
.
M
i
sh
r
a,
"C
o
n
t
i
n
u
o
u
s
P
u
n
j
ab
i
sp
eech
r
eco
g
n
i
t
i
o
n
m
o
d
el
b
ased
o
n
K
al
d
i
A
S
R
to
o
lk
it"
,
In
t
e
rn
a
t
i
o
n
a
l
J
o
u
rn
a
l
of
S
pe
e
c
h
T
e
c
hnol
ogy,
2018.
[1
2
]
Ta
c
h
i
c
a
r
t
R
i
d
o
u
a
n
e
,
K
a
r
im
M
o
h
a
m
e
d
B
o
u
z
o
u
b
a
a
,
S
i
L
h
ous
s
a
i
n
A
our
a
gh,
H
a
m
i
d
J
a
a
f
a
r
.
"A
u
t
o
m
at
i
c
I
d
en
t
i
f
i
cat
i
o
n
of
M
or
oc
c
a
n
C
ol
l
oqui
a
l
A
r
a
bi
c
"
. J
a
n
u
a
r
y
2
0
1
8
in
b
o
o
k
Ar
a
b
i
c
L
a
n
g
u
a
g
e
P
r
o
c
e
s
s
i
n
g
:
F
r
o
m
T
h
e
o
r
y
t
o
P
r
a
c
t
i
c
e
.
[1
3
]
Af
i
f
y
,
M
.
,
S
a
r
i
k
a
y
a
,
R
.
,
k
wa
n
g
J
e
f
f
Ku
o
,
H.
,
B
e
s
a
c
i
e
r
,
L
.
,
Ga
o
,
Y.
,
2
0
0
6
.
"
On
t
h
e
u
s
e
o
f
mo
r
p
h
o
l
o
g
i
c
a
l
a
n
a
l
y
s
i
s
f
o
r
di
a
l
e
c
t
a
l
a
r
a
bi
c
s
pe
e
c
h
r
e
c
ogni
t
i
on"
,
i
n:
P
r
oc
e
e
di
ngs
of
I
C
S
L
P
06,
pp.
277
–
280.
[1
4
]
Al
i
A.
,
De
h
a
k
N.
,
C
a
r
d
i
n
a
l
P
.
,
Kh
u
r
a
n
a
S
.
,
Ye
l
l
a
S
.
,
Gl
a
s
s
J
.
,
B
e
l
l
P
.
,
R
e
n
a
l
s
S
.
,
2
0
1
6
.
"
Au
t
o
ma
tic
d
ia
le
c
t
d
e
te
c
tio
n
in
a
r
a
b
ic
b
r
o
a
d
c
a
s
t s
p
e
e
c
h
"
. C
o
R
R
0
8
-
12
-
Se
p
t
e
m
b
e
r
-
2016,
2934
–
2938.
doi
:
10.
21437/
I
nt
e
r
s
pe
e
c
h.
2016
-
1297.
[1
5
]
Al
i
,
A.
,
M
u
b
a
r
a
k
,
H.
,
Vo
g
e
l
,
S
.
,
2
0
1
4
a
.
"A
d
v
an
ces
i
n
d
i
al
ect
al
ar
ab
i
c
sp
eech
r
eco
g
n
i
t
i
o
n
:
A
st
u
d
y
u
si
n
g
t
w
i
t
t
er
t
o
im
p
r
o
v
e
e
g
y
p
tia
n
AS
R
"
,
i
n
:
I
n
t
e
r
n
a
t
i
o
n
a
l
W
o
r
k
s
h
o
p
o
n
S
p
o
k
e
n
L
a
n
g
u
a
g
e
T
r
a
n
s
l
a
t
i
o
n
(
I
W
S
L
T
2
0
1
4
)
.
[1
6
]
El
m
a
h
d
y
,
M
.
,
H
a
s
e
g
a
w
a
-
Jo
h
n
so
n
,
M
.
,
M
u
st
af
aw
i
,
E
.
,
2
0
1
2
.
"A
b
asel
i
n
e
sp
eech
r
eco
g
n
i
t
i
o
n
sy
st
em
f
o
r
l
ev
an
t
i
n
e
co
l
l
o
q
u
i
al
ar
ab
i
c".
P
r
o
ceed
i
n
g
s
o
f
E
S
O
L
E
C
.
[1
7
]
El
m
a
h
d
y
,
M
.
,
H
a
s
e
g
a
w
a
-
Jo
h
n
so
n
,
M
.
,
M
u
st
af
aw
i
,
E
.
,
2
0
1
4
.
"D
ev
el
o
p
m
en
t
o
f
a
t
v
b
r
o
ad
cast
s
sp
eech
r
eco
g
n
i
t
i
o
n
sy
st
em
f
o
r
q
at
ar
i
ar
ab
i
c",
i
n
:
L
R
E
C
,
p
p
.
3
0
5
7
–
3061.
[1
8
]
So
l
t
a
u
,
H
.
,
M
a
n
g
u
,
L.
,
B
i
a
d
s
y
,
F.
,
2
0
1
1
.
"F
r
o
m
m
o
d
er
n
st
an
d
ar
d
ar
ab
i
c
t
o
l
ev
an
t
i
n
e
A
S
R
:
L
ev
er
ag
i
ng
ga
l
e
f
or
di
a
l
e
c
t
s
"
,
i
n:
2011
I
E
E
E
W
or
ks
hop
on
A
ut
om
a
t
i
c
S
pe
e
c
h
R
e
c
ogni
t
i
on
U
nde
r
s
t
a
ndi
ng,
pp.
266
–
271.
doi
:
10.
1109/
A
S
R
U
.
2011.
6163942.
[1
9
]
Ve
r
g
y
r
i
,
D.
,
Ki
r
c
h
h
o
f
f
,
K.
,
Ga
d
d
e
,
V.
R
.
R
.
,
S
t
o
l
c
k
e
,
A.
,
Z
h
e
n
g
,
J
.
,
2
0
0
5
.
"
De
v
e
l
o
p
me
n
t
o
f
a
c
o
n
v
e
r
s
a
t
i
o
n
a
l
te
le
p
h
o
n
e
s
p
e
ech
r
eco
g
n
i
zer
f
o
r
l
ev
an
t
i
n
e
ar
ab
i
c",
i
n
:
P
r
o
ceed
i
n
g
s
o
f
I
n
t
er
sp
eech
,
C
i
t
eseer
.
p
p
.
1
6
1
3
–
1616.
[2
0
]
Mo
u
a
z
B
e
z
o
u
i
,
A
b
d
e
r
r
a
h
i
m
B
e
n
i
-
hs
s
a
ne
a
nd
A
bde
l
m
a
j
i
d
E
l
m
out
a
oua
kki
l
,
"
Sp
e
e
c
h
R
e
c
o
g
n
i
t
i
o
n
o
f
M
o
r
o
c
can
Di
a
l
e
c
t
Us
i
n
g
Hi
d
d
e
n
M
a
r
k
o
v
M
o
d
e
l
s
"
In
t
e
rn
a
t
i
o
n
a
l
S
y
m
p
o
s
i
u
m
o
n
M
a
c
h
i
n
e
L
e
a
rn
i
n
g
a
n
d
B
i
g
D
a
t
a
A
n
a
l
y
t
i
c
s
fo
r
Cy
b
e
r
s
e
c
u
r
i
t
y
a
n
d
P
r
i
v
a
c
y
(
M
L
BD
A
CP
)
A
p
r
i
l
2
9
–
Ma
y
2
,
2
0
1
9
,
L
e
u
v
e
n
,
B
e
l
g
i
u
m
.
[2
1
]
Wr
a
y
,
S
.
,
A
l
i
,
A
.
,
2
0
1
5
.
"
C
r
o
w
d
s
o
u
r
c
e
a
l
i
t
t
l
e
t
o
l
a
b
e
l
a
l
o
t
:
L
a
b
e
lin
g
a
s
p
e
e
c
h
c
o
r
p
u
s
o
f
d
ia
le
c
ta
l
A
r
a
b
ic
"
.
In
t
e
rn
a
t
i
o
n
a
l
S
p
e
e
c
h
a
n
d
C
o
m
m
u
n
i
-
cat
i
o
n
A
sso
ci
at
i
o
n
.
v
o
l
u
m
e
2
0
1
5
-
Jan
u
ar
y
.
p
p
.
2
8
2
4
–
2828.
BI
O
G
R
A
P
H
IE
S
OF
A
U
T
HOR
S
Be
z
o
u
i
Mo
u
a
z
(b
o
rn
i
n
1
9
8
5
)
re
c
e
i
v
e
d
a
m
a
s
t
e
r’s
d
e
g
re
e
o
f
n
e
t
w
o
rk
s
a
n
d
t
e
l
e
c
o
m
m
u
n
i
c
a
t
i
o
n
s
fro
m
C
h
o
u
a
i
b
D
o
u
k
k
a
l
i
U
n
i
v
e
rs
i
t
y
,
F
a
c
u
l
t
y
o
f
S
c
i
e
n
c
e
E
I
J
a
d
i
d
a
,
M
o
ro
c
c
o
i
n
2
0
1
1
.
H
e
i
s
cu
r
r
en
t
l
y
p
u
r
su
i
n
g
h
i
s
P
h
.
D
.
d
eg
r
ee
(
C
o
m
p
u
t
er
S
ci
en
ce)
at
t
h
e
C
h
o
u
ai
b
D
o
u
k
k
al
i
U
n
i
v
er
si
t
y
F
acu
l
t
y
o
f
S
ci
en
ce,
E
l
Jad
i
d
a,
M
o
r
o
cco
.
H
i
s
r
esear
ch
i
n
t
er
est
s
i
n
cl
u
d
e
S
p
eech
R
eco
g
n
i
t
i
o
n
an
d
Si
g
n
a
l
Pr
o
c
e
s
s
i
n
g
.
Be
n
i
-
hs
s
a
ne
A
bde
r
r
a
hi
m
is
c
u
r
r
e
n
tly
a
p
r
o
f
e
s
s
o
r
in
C
h
o
u
a
ib
D
o
u
k
k
a
li
Un
i
v
e
r
s
i
t
y
F
a
c
u
l
t
y
o
f
Sc
i
e
n
c
e
,
D
e
p
a
r
t
m
e
n
t
o
f
C
o
m
p
u
t
e
r
s
c
i
e
n
c
e
,
LA
R
O
SER
I
La
b
o
r
a
t
o
r
y
,
EI
J
a
d
i
d
a
,
M
o
r
o
c
c
o
.
H
i
s
re
s
e
a
rc
h
i
n
t
e
re
s
t
s
i
n
c
l
u
d
e
D
a
t
a
M
i
n
i
n
g
a
n
d
K
n
o
w
l
e
d
g
e
D
i
s
c
o
v
e
ry
.
El
m
o
u
t
a
o
u
a
k
k
i
l
A
bde
l
m
a
j
i
d
is
c
u
r
r
e
n
tly
a
p
r
o
f
e
s
s
o
r
in
Ch
o
u
a
i
b
D
o
u
k
k
a
l
i
U
n
i
v
e
r
s
i
t
y
F
a
c
u
l
t
y
o
f
Sc
i
e
n
c
e
,
D
e
p
a
r
t
m
e
n
t
o
f
C
o
m
p
u
t
e
r
s
c
i
e
n
c
e
,
LA
R
O
SER
I
La
b
o
r
a
t
o
r
y
,
EI
J
a
d
i
d
a
,
M
o
r
o
c
c
o
.
H
i
s
re
s
e
a
rc
h
i
n
t
e
re
s
t
s
i
n
c
l
u
d
e
M
e
d
i
c
a
l
Im
a
g
e
P
ro
c
e
s
s
i
n
g
a
n
d
S
p
e
e
c
h
R
e
c
o
g
n
i
t
i
o
n
.
Evaluation Warning : The document was created with Spire.PDF for Python.