I
AE
S
I
nte
rna
t
io
na
l J
o
urna
l o
f
Art
if
icia
l In
t
ellig
ence
(
I
J
-
AI
)
Vo
l.
9
,
No
.
3
,
Sep
tem
b
er
2020
,
p
p
.
387
~
3
93
I
SS
N:
2252
-
8938
,
DOI
: 1
0
.
1
1
5
9
1
/i
j
ai.
v
9
.i
3
.
p
p
387
-
3
93
387
J
o
ur
na
l ho
m
ep
a
g
e
:
h
ttp
:
//ij
a
i
.
ia
esco
r
e.
co
m
Co
m
pa
riso
n of
C
NNs a
nd S
VM
f
o
r vo
ice control w
heelcha
ir
M
o
ha
m
m
a
d S
ha
hrul
I
zha
m
Sh
a
rif
ud
di
n,
Sh
a
rif
a
lill
a
h N
o
rdin
,
Azliza
M
o
hd
Ali
F
a
c
u
lt
y
o
f
Co
m
p
u
ter an
d
M
a
th
e
m
a
ti
c
a
l
S
c
ien
c
e
s,
Un
iv
e
rsiti
T
e
k
n
o
lo
g
i
M
A
R
A
,
S
h
a
h
A
la
m
,
S
e
lan
g
o
r,
M
a
lay
sia
Art
icle
I
nfo
AB
ST
RAC
T
A
r
ticle
his
to
r
y:
R
ec
eiv
ed
F
eb
10
,
2
0
20
R
ev
i
s
ed
A
pr
2
0
,
2
0
20
A
cc
ep
ted
M
ay
6
,
2
0
20
In
t
h
is
p
a
p
e
r,
w
e
d
e
v
e
lo
p
a
n
in
t
e
ll
ig
e
n
t
w
h
e
e
lch
a
ir
u
sin
g
CNN
s
a
n
d
S
VM
v
o
ice
re
c
o
g
n
it
io
n
m
e
th
o
d
s.
T
h
e
d
a
ta
is
c
o
ll
e
c
ted
f
ro
m
G
o
o
g
le
a
n
d
so
m
e
o
f
th
e
m
a
re
s
e
lf
-
r
e
c
o
rd
e
d
.
T
h
e
re
a
re
f
o
u
r
ty
p
e
s
o
f
d
a
ta
to
b
e
re
c
o
g
n
i
z
e
d
w
h
ich
a
re
g
o
,
lef
t,
ri
g
h
t,
a
n
d
sto
p
.
Vo
i
c
e
d
a
ta
a
re
e
x
tra
c
ted
u
sin
g
M
F
CC
f
e
a
tu
re
e
x
trac
ti
o
n
tec
h
n
i
q
u
e
.
CNN
s
a
n
d
S
V
M
a
re
th
e
n
u
se
d
t
o
c
las
sify
a
n
d
re
c
o
g
n
ize
th
e
v
o
ice
d
a
ta.
T
h
e
m
o
to
r
d
riv
e
r
i
s
e
m
b
e
d
d
e
d
in
Ra
sp
b
e
rry
P
I
3
B+
to
c
o
n
tr
o
l
th
e
m
o
v
e
m
e
n
t
o
f
th
e
w
h
e
e
lch
a
ir
p
ro
t
o
ty
p
e
.
CNN
s
p
ro
d
u
c
e
d
h
ig
h
e
r
a
c
c
u
ra
c
y
i.
e
.
9
5
.
3
0
%
c
o
m
p
a
re
d
to
S
V
M
w
h
ich
is
o
n
ly
7
2
.
3
9
%
.
On
t
h
e
o
th
e
r
h
a
n
d
,
S
V
M
o
n
ly
to
o
k
8
.
2
1
se
c
o
n
d
s
w
h
il
e
CNN
s
to
o
k
2
5
0
.
0
3
se
c
o
n
d
s
to
e
x
e
c
u
te.
T
h
e
re
f
o
re
,
CNN
s
p
ro
d
u
c
e
b
e
tt
e
r
re
su
lt
b
e
c
a
u
se
n
o
ise
a
re
f
il
tere
d
in
th
e
f
e
a
tu
re
e
x
tra
c
ti
o
n
lay
e
r
b
e
f
o
re
c
la
ss
if
ied
in
th
e
c
las
si
f
ic
a
ti
o
n
lay
e
r.
Ho
w
e
v
e
r,
CNN
s
to
o
k
lo
n
g
e
r
ti
m
e
d
u
e
to
t
h
e
c
o
m
p
lex
it
y
o
f
th
e
n
e
tw
o
rk
s
a
n
d
t
h
e
les
s
c
o
m
p
lex
it
y
i
m
p
le
m
e
n
tatio
n
i
n
S
VM
g
iv
e
sh
o
rter p
r
o
c
e
ss
in
g
ti
m
e
.
K
ey
w
o
r
d
s
:
C
o
n
v
o
lu
tio
n
al
n
e
u
r
al
n
e
t
w
o
r
k
s
Su
p
p
o
r
t v
ec
to
r
m
ac
h
i
n
e
Vo
ice
r
ec
o
g
n
itio
n
T
h
is i
s
a
n
o
p
e
n
a
c
c
e
ss
a
rticle
u
n
d
e
r th
e
CC B
Y
-
SA
li
c
e
n
se
.
C
o
r
r
e
s
p
o
nd
ing
A
uth
o
r
:
Sh
ar
i
f
alilla
h
No
r
d
in
,
Facu
lt
y
o
f
C
o
m
p
u
ter
an
d
Ma
t
h
e
m
a
tical
Scie
n
ce
s
,
Un
i
v
er
s
iti T
ek
n
o
lo
g
i
M
A
R
A
,
Sh
a
h
A
la
m
,
Selan
g
o
r
,
Ma
la
y
s
i
a
.
E
m
ail: s
h
ar
i
f
a
@
f
s
k
m
.
u
it
m
.
ed
u
.
m
y
1.
I
NT
RO
D
UCT
I
O
N
I
m
p
air
e
d
p
e
o
p
le
is
in
c
r
ea
s
in
g
ev
er
y
y
ea
r
d
u
e
to
ag
ein
g
,
ac
ci
d
en
ts
an
d
als
o
d
is
e
ases
s
u
ch
a
s
p
a
r
a
ly
s
is
,
s
p
in
al
co
r
d
in
j
u
r
i
es
(
SC
I
)
,
am
p
u
tat
io
n
an
d
q
u
a
d
r
i
p
l
eg
ia
[
1
]
.
T
h
is
g
r
o
u
p
o
f
p
e
o
p
le
r
e
q
u
ir
e
a
s
p
ec
ial
w
h
ee
lch
ai
r
to
m
o
v
e
s
u
ch
as
a
m
o
to
r
is
ed
w
h
ee
lch
ai
r
.
M
o
t
o
r
is
e
d
w
h
ee
lc
h
air
c
o
n
s
is
ts
o
f
el
ec
t
r
i
c
m
o
to
r
an
d
jo
y
s
tick
at
th
e
ar
m
r
est
.
Ho
w
ev
er
,
d
is
a
b
le
d
p
eo
p
l
e
w
ith
p
ar
aly
s
ed
h
an
d
o
r
h
av
in
g
a
p
r
o
b
lem
w
ith
th
e
h
an
d
c
an
n
o
t
o
p
e
r
at
e
jo
y
s
tick
o
n
th
e
m
o
to
r
is
e
d
w
h
ee
l
ch
ai
r
t
o
m
o
v
e.
T
h
er
ef
o
r
e
,
a
l
o
t
o
f
r
es
ea
r
ch
es
s
tu
d
ie
d
t
o
h
elp
th
e
im
p
ai
r
e
d
p
e
o
p
le
m
o
v
e
th
e
w
h
ee
lch
air
-
u
s
in
g
v
o
ic
e
r
e
c
o
g
n
iti
o
n
[1
-
3]
.
T
h
is
w
h
ee
lch
ai
r
ca
n
b
e
c
all
ed
s
m
ar
t
o
r
in
te
llig
en
t
w
h
ee
lch
ai
r
b
e
ca
u
s
e
it
c
an
b
e
m
o
v
ed
u
s
in
g
o
n
l
y
th
e
v
o
ice.
T
h
e
f
ir
s
t
in
tell
ig
en
t
w
h
ee
lch
ai
r
w
as
d
ev
el
o
p
ed
b
y
th
e
Siam
o
Un
iv
e
r
s
ity
o
f
A
lcal
a,
S
p
ain
in
1
9
9
9
[
4
]
.
T
h
e
im
p
r
o
v
em
en
t
o
f
v
o
i
ce
an
d
s
p
e
ec
h
i
d
en
t
if
ica
ti
o
n
t
ec
h
n
o
l
o
g
y
h
as
b
e
en
s
ta
r
te
d
b
y
T
ex
as
I
n
s
tr
u
m
en
ts
in
1
9
6
0
[
5
]
.
R
e
ce
n
tly
,
v
o
ice
r
ec
o
g
n
iti
o
n
o
r
s
p
ee
ch
r
ec
o
g
n
iti
o
n
h
as b
ee
n
a
p
p
li
e
d
in
ass
is
tin
g
p
e
o
p
le
d
o
in
g
w
o
r
k
th
r
o
u
g
h
d
ig
i
tal
d
e
v
ices
s
u
ch
as
m
o
b
ile
p
h
o
n
e,
t
a
b
le
ts
,
an
d
p
e
r
s
o
n
al
co
m
p
u
te
r
.
T
h
er
e
a
r
e
als
o
v
o
i
c
e
r
e
co
g
n
it
io
n
s
o
f
tw
ar
e
an
d
w
eb
p
ag
es
i
.
e
.
g
o
o
g
le
a
p
p
lic
ati
o
n
s
,
tr
an
s
la
ti
o
n
s
o
f
tw
ar
e,
an
d
p
e
r
s
o
n
al
ass
is
tan
ts
s
u
ch
as A
lex
a,
C
o
r
t
an
a
an
d
S
ir
i.
T
h
i
s
p
e
r
s
o
n
al
ass
is
t
an
t
ca
n
b
e
ca
ll
ed
m
o
d
e
r
n
ch
a
tb
o
t
w
h
ich
ca
n
ass
is
t
p
e
o
p
l
e
in
an
y
to
p
ics
th
ey
w
o
u
ld
l
ik
e
to
ask
.
T
h
e
ad
v
an
ce
m
en
t
o
f
tec
h
n
o
lo
g
y
in
ar
ti
f
icial
i
n
telli
g
e
n
ce
(
A
I
)
h
av
e
cr
ea
ted
a
s
m
ar
t
w
h
ee
lch
air
o
r
in
telli
g
e
n
t
w
h
ee
lch
air
e
x
p
lo
itin
g
t
h
e
v
o
ice
r
ec
o
g
n
itio
n
f
ea
tu
r
es.
A
lo
t
o
f
r
esear
ch
es
h
av
e
b
ee
n
co
n
d
u
cted
to
i
m
p
r
o
v
e
t
h
e
f
u
n
ctio
n
ali
t
y
o
f
t
h
e
w
h
ee
lc
h
air
.
Na
s
r
in
[
1
]
pr
o
p
o
s
ed
an
ap
p
licatio
n
o
n
s
m
ar
t
w
h
ee
lch
air
-
u
s
i
n
g
v
o
ice
an
d
ad
d
GP
S
f
u
n
ctio
n
to
tr
ac
k
t
h
e
u
s
er
n
a
v
ig
a
tio
n
a
n
d
lo
ca
tio
n
.
T
h
is
ap
p
lica
tio
n
r
eq
u
ir
ed
W
I
FI
co
n
n
ec
tio
n
a
n
d
t
h
e
ap
p
licatio
n
i
s
i
n
s
ta
lled
i
n
a
m
o
b
ile
e
n
v
ir
o
n
m
e
n
t.
Ho
w
e
v
er
,
Av
u
t
u
et.
al.
[
3
]
p
r
o
p
o
s
ed
a
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
2
5
2
-
8938
I
n
t J
A
r
ti
f
I
n
tell
,
Vo
l.
9
,
No
.
3
,
Sep
te
m
b
er
20
20
:
3
8
7
–
3
93
388
lo
w
-
co
s
t
m
ap
f
o
r
v
o
ice
r
ec
o
g
n
itio
n
w
h
ee
lc
h
air
w
h
ic
h
ca
n
b
e
ap
p
lied
in
a
l
o
ca
l
lo
ca
tio
n
.
T
h
is
ap
p
licatio
n
ca
n
b
e
u
s
ed
w
i
th
o
u
t
a
n
y
n
e
t
w
o
r
k
co
n
n
ec
t
io
n
.
T
h
e
co
s
t
i
s
m
u
c
h
c
h
ea
p
er
co
m
p
ar
ed
to
th
e
a
p
p
licatio
n
u
s
i
n
g
a
n
et
w
o
r
k
co
n
n
ec
t
io
n
.
B
ar
r
iu
s
o
[
6
]
p
r
o
p
o
s
ed
a
s
m
ar
tp
h
o
n
e
a
p
p
licatio
n
f
o
r
a
w
h
ee
lc
h
air
with
a
n
a
g
en
t
-
b
ased
in
telli
g
e
n
t
i
n
ter
f
ac
e.
An
o
t
h
er
ap
p
licatio
n
w
a
s
d
ev
el
o
p
ed
u
s
in
g
A
r
d
u
i
n
o
Me
g
a
2
5
6
0
b
y
[
7
]
.
T
h
e
s
p
ec
ial
f
ea
tu
r
es
i
n
th
i
s
ap
p
licatio
n
ar
e
th
e
e
m
er
g
en
c
y
m
ess
a
g
es
ca
n
b
e
s
en
t
to
th
e
i
m
p
o
r
tan
t
p
eo
p
le.
W
ith
th
e
u
ltra
s
o
n
ic
s
e
n
s
o
r
,
a
n
y
o
b
s
tacle
ca
n
b
e
d
etec
ted
a
n
d
av
o
id
ed
w
ith
o
u
t
i
n
ter
n
et
c
o
n
n
ec
tio
n
r
eq
u
ir
ed
.
T
h
er
e
is
also
an
ap
p
licatio
n
o
f
i
n
telli
g
e
n
t
w
h
ee
lc
h
air
f
o
r
p
eo
p
le
w
it
h
p
ar
ap
leg
ia
p
r
o
b
lem
p
r
o
p
o
s
ed
b
y
[
8
]
.
T
h
e
ap
p
li
ca
tio
n
ap
p
lied
v
o
i
ce
r
ec
o
g
n
itio
n
a
n
d
to
u
c
h
s
cr
ee
n
co
n
tr
o
l.
Me
an
w
h
ile,
C
h
a
u
h
an
[
9
]
d
id
a
co
m
p
ar
ati
v
e
s
tu
d
y
o
n
v
o
ice
r
e
co
g
n
itio
n
w
h
ee
lc
h
air
an
d
p
r
o
p
o
s
ed
a
n
e
w
d
esi
g
n
d
ep
lo
y
i
n
g
i
n
f
r
ar
ed
s
e
n
s
o
r
a
n
d
R
asb
er
r
y
P
i th
r
o
u
g
h
U
SB
m
ic
r
o
p
h
o
n
e.
T
h
e
d
ev
e
lo
p
m
en
t
o
f
s
m
ar
t
o
r
in
tel
lig
en
t
w
h
ee
lch
ai
r
m
o
s
tly
b
as
e
d
o
n
A
I
t
ec
h
n
o
lo
g
i
es.
T
h
e
p
r
o
ce
s
s
o
f
v
o
i
ce
r
e
co
g
n
it
io
n
ap
p
licatio
n
w
ill
b
eg
in
ex
tr
ac
tin
g
f
ea
tu
r
es
an
d
cl
ass
if
ic
ati
o
n
.
T
h
er
e
a
r
e
a
l
o
t
o
f
f
ea
tu
r
es
ex
t
r
a
cti
o
n
t
ec
h
n
i
q
u
e
f
o
r
v
o
i
ce
r
ec
o
g
n
iti
o
n
s
u
ch
as
L
in
ea
r
Pre
d
i
cti
o
n
C
o
ef
f
ici
en
ts
(
L
P
C
)
,
Dis
c
r
et
e
W
av
e
let
T
r
an
s
f
o
r
m
(
D
W
T
)
,
L
in
e
S
p
ec
t
r
a
l
F
r
e
q
u
en
ci
es
(
L
SF
)
,
an
d
M
el
-
F
r
eq
u
en
cy
C
ep
s
tr
u
m
C
o
ef
f
ici
en
ts
(
MFC
C
)
[
1
0
]
.
C
lass
if
ic
ati
o
n
tech
n
i
q
u
es
c
an
b
e
a
p
p
li
ed
in
c
lass
if
y
in
g
an
d
r
e
c
o
g
n
is
in
g
th
e
v
o
ic
e,
es
p
e
cia
lly
in
w
h
ee
lch
ai
r
ap
p
l
ic
ati
o
n
.
T
h
e
m
o
s
t
p
o
p
u
l
a
r
class
if
ica
ti
o
n
te
ch
n
iq
u
es
an
d
p
r
o
m
is
in
g
w
ith
h
ig
h
ac
cu
r
ac
y
ar
e
co
n
v
o
lu
ti
o
n
al
n
eu
r
al
n
e
tw
o
r
k
s
(
C
NNs)
.
C
NNs
is
b
as
e
d
o
n
f
ee
d
f
o
r
w
ar
d
n
eu
r
al
n
e
tw
o
r
k
an
d
h
av
e
b
ett
e
r
g
en
er
a
lis
a
ti
o
n
th
an
n
etw
o
r
k
s
w
ith
f
u
ll
co
n
n
ec
tiv
ity
b
etw
ee
n
ad
j
a
ce
n
t
l
ay
er
s
[
1
1
]
.
Mo
r
e
o
v
er
,
C
NNs
h
as
b
e
en
s
u
cc
ess
f
u
lly
ap
p
li
ed
in
m
an
y
ap
p
li
ca
t
io
n
s
esp
ec
i
ally
in
im
ag
e
id
en
tif
i
ca
t
io
n
[
1
2
-
1
5
]
,
s
u
r
v
eill
an
ce
[
1
6
]
,
an
d
h
u
m
an
r
e
co
g
n
it
io
n
[
1
7
-
1
9
]
.
L
ei
an
d
Sh
e
[
2
0
]
au
th
en
tic
ate
v
o
ice
u
s
in
g
C
NNs
m
et
h
o
d
in
a
n
o
i
s
y
en
v
ir
o
n
m
e
n
t.
T
h
eir
r
esear
ch
p
r
o
d
u
ce
s
b
etter
ac
cu
r
ac
y
an
d
r
ed
u
ce
s
an
eq
u
al
er
r
o
r
r
ate.
Gu
a
n
[
2
1
]
o
p
tim
is
e
p
er
f
o
r
m
a
n
ce
in
s
p
ee
c
h
r
ec
o
g
n
itio
n
u
s
in
g
C
NN
s
an
d
th
e
r
ec
o
g
n
itio
n
p
er
f
o
r
m
an
ce
h
as
in
cr
ea
s
ed
w
it
h
an
er
r
o
r
r
ate
o
f
1
3
.
8
8
%.
Sh
ar
i
f
u
d
d
i
n
e
t.a
l
co
m
p
ar
ed
C
N
N
s
a
n
d
B
P
NNs
o
n
v
o
ice
co
n
tr
o
l
w
h
ee
lc
h
air
ap
p
licatio
n
s
a
n
d
p
r
o
d
u
ce
s
h
ig
h
ac
cu
r
ac
y
o
n
C
NNs tec
h
n
iq
u
e
[
5
]
.
Su
p
p
o
r
t
Ve
ct
o
r
M
ac
h
in
e
(
SV
M)
c
an
b
e
c
ateg
o
r
is
e
d
as
o
n
e
o
f
th
e
b
est
cl
ass
if
ic
ati
o
n
s
t
ec
h
n
iq
u
es.
SVM
als
o
h
as
b
e
en
a
p
p
lie
d
in
v
ar
i
o
u
s
a
p
p
lic
ati
o
n
s
f
o
r
in
s
t
a
n
ce
in
b
io
m
etr
ics
[
1
7
,
2
2
-
2
3
]
,
s
en
tim
en
t
an
aly
s
is
[
2
4
-
2
5
]
an
d
s
ec
u
r
ity
s
u
ch
as
in
tr
u
s
i
o
n
d
et
ec
t
io
n
[
2
6
-
2
7
]
.
Selv
ak
u
m
ar
i
an
d
R
a
d
h
a
[
2
8
]
a
p
p
li
e
d
SVM
in
class
if
y
in
g
s
p
e
ec
h
p
a
th
o
l
o
g
y
an
d
a
ch
iev
e
d
9
8
%
ac
cu
r
a
cy
co
m
p
ar
ed
t
o
th
e
N
aïv
e
B
ay
es
alg
o
r
i
th
m
.
Fu
r
th
e
r
m
o
r
e
,
A
s
tu
ti
an
d
R
iy
an
d
w
ita
[
2
9
]
a
p
p
ly
in
g
SVM
in
r
ec
o
g
n
is
in
g
v
o
ic
e
f
o
r
s
t
a
r
tin
g
a
ca
r
en
g
in
e
.
SVM
cl
ass
if
ie
d
th
e
w
o
r
d
‘
o
n
’
an
d
‘
o
f
f
’
ca
r
en
g
in
e
w
ith
9
2
.
1
5
%
ac
cu
r
ac
y
.
Me
an
w
h
ile,
H
ar
v
i
an
to
et
.
al
[
3
0
]
an
aly
s
e
th
e
v
o
i
ce
o
f
I
n
d
o
n
esi
a
L
an
g
u
ag
e
u
s
in
g
MFC
C
an
d
S
VM
ac
h
iev
ed
h
ig
h
a
cc
u
r
a
cy
w
h
ich
is
9
1
.
8
3
%
.
I
n
th
is
p
a
p
e
r
,
w
e
p
r
o
p
o
s
e
an
i
n
tellig
en
t
w
h
ee
lch
air
-
u
s
in
g
v
o
ice
r
e
c
o
g
n
iti
o
n
.
T
h
er
e
a
r
e
f
o
u
r
ty
p
es
o
f
v
o
ic
e
co
m
m
an
d
to
b
e
r
e
co
g
n
i
ze
d
w
h
ich
ar
e
g
o
,
lef
t
,
r
ig
h
t
an
d
s
to
p
.
T
h
e
d
ata
is
c
o
ll
ec
te
d
f
r
o
m
Go
o
g
l
e
an
d
s
o
m
e
o
f
th
em
ar
e
s
elf
-
r
ec
o
r
d
e
d
.
Fe
atu
r
es
f
r
o
m
th
e
d
ata
a
r
e
e
x
tr
a
cte
d
u
s
in
g
MFC
C
t
ec
h
n
i
q
u
es.
Fo
ll
o
w
ed
b
y
th
e
r
e
co
g
n
it
io
n
u
s
in
g
C
NNs.
W
e
a
ls
o
class
if
y
th
e
v
o
ice
u
s
in
g
th
e
SVM
m
eth
o
d
to
c
o
m
p
a
r
e
th
e
ef
f
icien
cy
an
d
ac
cu
r
a
cy
b
etw
ee
n
th
e
tw
o
tec
h
n
iq
u
e
s
.
T
h
is
p
ap
er
is
o
r
g
an
i
s
e
d
as f
o
llo
w
s
.
Sect
io
n
2
p
r
esen
t
s
r
esear
ch
m
et
h
o
d
s
o
n
v
o
ice
r
ec
o
g
n
i
tio
n
,
C
NN
s
an
d
SVM.
E
x
p
er
i
m
e
n
tat
io
n
d
etails
a
n
d
r
es
u
lt
s
w
it
h
th
e
co
m
p
ar
i
s
o
n
s
b
et
w
ee
n
C
NNs a
n
d
SVM
ar
e
p
r
esen
ted
in
s
ec
tio
n
3
.
Sectio
n
4
d
is
cu
s
s
es th
e
co
n
cl
u
s
io
n
o
f
t
h
e
p
ap
er
.
2.
RE
S
E
ARCH
M
E
T
H
O
D
T
h
is
p
a
p
e
r
f
o
cu
s
es
o
n
f
iv
e
ty
p
es
o
f
v
o
ic
e
d
a
ta
.
Sp
e
ec
h
s
ig
n
a
l
is
b
ase
d
o
n
s
in
g
le
w
o
r
d
s
;
s
p
ea
k
e
r
s
ar
e
in
d
e
p
en
d
en
t
;
lan
g
u
ag
e
is
E
n
g
lis
h
;
v
o
c
a
b
u
la
r
y
ty
p
e
is
s
m
all,
an
d
b
a
ck
g
r
o
u
n
d
n
o
is
e
is
t
r
u
e
.
T
a
b
le
1
p
r
es
en
ts
th
e
f
iv
e
ty
p
es
o
f
v
o
i
ce
d
ata
u
s
e
d
in
th
is
p
ap
er
.
T
ab
le
1
.
T
y
p
es o
f
co
llected
v
o
ice
d
ata
Ty
p
e
D
e
scri
p
t
i
o
n
S
p
e
e
c
h
si
g
n
a
l
S
i
n
g
l
e
w
o
r
d
s
S
p
e
a
k
e
r
I
n
d
i
v
i
d
u
a
l
L
a
n
g
u
a
g
e
En
g
l
i
sh
V
o
c
a
b
u
l
a
r
y
S
mal
l
B
a
c
k
g
r
o
u
n
d
n
o
i
se
Y
e
s
Dev
el
o
p
m
en
t
b
o
a
r
d
c
o
n
ta
in
s
m
icr
o
p
r
o
c
ess
o
r
an
d
m
icr
o
c
o
n
t
r
o
lle
r
in
a
s
m
all,
c
o
m
p
ac
t
c
ir
cu
it
b
o
ar
d
.
T
h
e
d
ev
el
o
p
m
en
t
b
o
a
r
d
r
ec
eiv
e
th
e
in
p
u
t
,
p
r
o
c
ess
th
e
c
o
m
m
an
d
,
an
d
p
r
o
d
u
c
e
th
e
o
u
tp
u
t
.
S
ev
er
al
d
ev
el
o
p
m
en
t
b
o
a
r
d
s
a
r
e
in
th
e
m
ar
k
et
w
ith
th
eir
s
t
r
en
g
th
s
an
d
lim
itati
o
n
s
,
f
o
r
ex
am
p
le
A
r
d
u
in
o
,
B
an
an
a
PI
,
an
d
R
as
p
b
er
r
y
Evaluation Warning : The document was created with Spire.PDF for Python.
I
n
t J
A
r
ti
f
I
n
tell
I
SS
N:
2252
-
8938
C
o
mp
a
r
is
o
n
o
f CN
N
s
a
n
d
S
V
M
fo
r
vo
ice
co
n
tr
o
l wh
ee
lch
a
i
r
(
Mo
h
a
mma
d
S
h
a
h
r
u
l I
z
h
a
m
S
h
a
r
ifu
d
d
in
)
389
PI.
R
as
p
b
e
r
r
y
P
I
3
B
+
w
ill
b
e
u
s
e
d
in
th
is
p
r
o
ject
d
u
e
to
th
e
f
o
ll
o
w
in
g
r
ea
s
o
n
s
;
f
as
t
p
r
o
ce
s
s
in
g
tim
e,
lo
w
co
s
t
an
d
ea
s
y
t
o
p
r
o
g
r
am
.
I
n
v
o
ice
r
ec
o
g
n
itio
n
,
C
NNs
i
s
s
ee
n
a
s
a
tec
h
n
iq
u
e
ca
p
ab
le
o
f
p
r
o
v
id
i
ng
h
ig
h
ac
c
u
r
ac
y
i
n
s
o
lv
i
n
g
esp
ec
iall
y
in
s
p
ee
c
h
a
n
d
i
m
a
g
e
r
ec
o
g
n
itio
n
p
r
o
b
le
m
s
[
2
0
]
.
C
NN
s
i
s
g
o
o
d
in
h
ig
h
li
g
h
t
i
n
g
i
m
p
o
r
tan
t
p
o
in
ts
,
as
w
e
ll a
s
f
ilter
i
n
g
n
o
is
e
i
n
th
e
d
ataset.
C
NNs
co
n
s
is
t o
f
f
ea
t
u
r
e
ex
tr
ac
tio
n
la
y
er
an
d
clas
s
i
f
icatio
n
la
y
er
.
2
.
1
.
F
e
a
t
ure
ex
t
ra
ct
io
n la
y
e
r
Featu
r
e
e
x
tr
ac
tio
n
la
y
er
co
n
tai
n
s
a
c
o
n
v
o
lu
t
io
n
al
l
a
y
er
a
n
d
p
o
o
lin
g
l
a
y
er
.
I
n
t
h
is
s
ec
tio
n
,
d
ata
w
ill b
e
ex
tr
ac
ted
b
ef
o
r
e
b
ee
n
f
ee
d
in
t
o
th
e
cla
s
s
i
f
icat
io
n
la
y
er
.
W
e
i
m
p
le
m
en
t
2
D
C
o
n
v
o
lu
tio
n
al
L
a
y
er
an
d
th
r
ee
2
D
Ma
x
P
o
o
lin
g
la
y
er
s
in
t
h
i
s
p
ap
er
.
2
.
2
.
Cla
s
s
if
ica
t
io
n la
y
er
C
las
s
i
f
icatio
n
la
y
er
ai
m
s
to
cl
ass
i
f
y
th
e
d
ata
r
ec
eiv
ed
f
r
o
m
t
h
e
f
ea
tu
r
e
e
x
tr
ac
tio
n
la
y
er
.
I
n
th
is
p
ap
er
,
th
e
clas
s
i
f
icatio
n
la
y
er
co
n
s
i
s
t
s
o
f
an
i
n
p
u
t la
y
er
,
a
h
id
d
en
la
y
er
,
an
d
an
o
u
tp
u
t la
y
er
.
On
th
e
o
t
h
er
h
an
d
,
th
is
p
ap
er
also
h
ig
h
l
ig
h
t
s
th
e
i
m
p
le
m
e
n
tatio
n
o
f
SVM
to
class
i
f
y
t
h
e
ex
tr
ac
ted
v
o
ice
d
ata.
SVM
is
ch
o
s
en
b
ec
au
s
e
o
f
t
h
e
ab
ilit
y
to
p
er
f
o
r
m
m
u
lti
-
clas
s
clas
s
if
ica
tio
n
w
i
th
h
i
g
h
d
i
m
en
s
io
n
al
s
p
ac
es.
T
h
r
ee
k
e
r
n
el
f
u
n
ctio
n
s
ar
e
u
s
ed
in
th
e
ex
p
er
i
m
e
n
t,
i.e
.
p
o
l
y
n
o
m
ial,
R
B
F,
an
d
s
i
g
m
o
id
.
W
e
th
en
co
n
d
u
c
t a
n
o
t
h
er
ex
p
e
r
i
m
en
t to
ca
lcu
late
t
h
e
ef
f
ec
t o
f
C
,
g
a
m
m
a,
an
d
d
eg
r
ee
u
s
i
n
g
th
e
k
er
n
el
r
es
u
lt.
Fig
u
r
e
1
p
r
esen
ts
t
h
e
f
lo
w
o
f
t
h
e
p
r
o
p
o
s
ed
s
y
s
te
m
[
5
]
.
W
e
d
escr
ib
ed
th
e
f
lo
w
b
ase
d
o
n
th
r
ee
m
o
d
u
les
:
th
e
i
n
p
u
t
(
th
e
u
s
er
i.
e.
th
e
d
is
ab
led
p
eo
p
le
)
,
th
e
p
r
o
ce
s
s
,
an
d
th
e
o
u
tp
u
t.
Firstl
y
,
th
e
u
s
er
p
u
s
h
e
s
th
e
b
u
tto
n
to
r
ec
o
r
d
th
e
v
o
ice
co
m
m
a
n
d
u
s
in
g
t
h
e
p
o
r
tab
le
m
icr
o
p
h
o
n
e.
MF
C
C
a
n
d
C
N
Ns
ar
e
u
s
ed
in
t
h
e
R
asp
b
er
r
y
P
I
3
B
+
to
p
r
o
ce
s
s
th
e
r
ec
o
r
d
ed
v
o
ic
e
.
T
h
e
o
u
tp
u
t
s
i
g
n
al
i
s
th
e
n
s
e
n
t
to
th
e
m
o
to
r
d
r
iv
er
in
th
e
R
asp
b
er
r
y
P
I
3
B
+
a
f
ter
th
e
s
y
s
te
m
r
ec
o
g
n
iz
e
d
th
e
v
o
ice.
T
h
e
m
o
to
r
d
r
iv
er
co
n
tr
o
ls
th
e
m
o
v
e
m
en
t
o
f
t
h
e
m
o
to
r
to
d
ir
ec
t
th
e
r
o
b
o
t c
ar
.
Fig
u
r
e
1
.
Flo
w
o
f
t
h
e
s
y
s
te
m
3.
E
XP
E
R
I
M
E
NT
A
T
I
O
NS
A
ND
RE
S
UL
T
S
3
.
1
.
Da
t
a
Sa
m
ple
T
h
er
e
ar
e
f
o
u
r
t
y
p
es
o
f
v
o
i
ce
co
m
m
an
d
u
s
ed
in
t
h
e
e
x
p
er
im
e
n
t.
T
h
ese
co
m
m
a
n
d
s
ar
e
r
ig
h
t,
lef
t,
s
to
p
,
an
d
g
o
.
T
h
e
d
ata
w
as
g
at
h
er
ed
f
r
o
m
Go
o
g
le.
T
h
er
e
ar
e
2
,
3
7
2
d
ata
f
o
r
g
o
co
m
m
an
d
,
2
,
3
8
0
f
o
r
s
to
p
co
m
m
a
n
d
,
2
,
3
5
3
f
o
r
lef
t
co
m
m
an
d
,
a
n
d
2
,
3
6
7
f
o
r
r
ig
h
t
co
m
m
an
d
.
T
h
er
e
ar
e
also
u
r
b
an
n
o
is
e
an
d
w
h
i
te
n
o
i
s
e
d
ata
co
llected
f
o
r
th
e
ex
p
er
i
m
en
t.
2
,
3
7
3
u
r
b
an
n
o
is
e
d
ata
ar
e
co
llected
f
r
o
m
Go
o
g
le
w
h
il
e
2
,
3
0
0
w
h
ite
n
o
is
e
d
ata
ar
e
s
elf
-
r
ec
o
r
d
ed
[
5
]
b
ec
au
s
e
t
h
e
d
ata
i
s
n
o
t
av
ai
lab
le
o
n
lin
e.
E
ac
h
v
o
ice
d
ata
is
p
r
e
p
ar
ed
in
o
n
e
-
s
ec
o
n
d
len
g
th
an
d
is
s
a
v
ed
in
W
A
V
f
o
r
m
at.
T
h
e
to
tal
n
u
m
b
er
o
f
d
o
w
n
lo
ad
ed
an
d
r
ec
o
r
d
e
d
v
o
i
c
e
co
m
m
an
d
u
s
ed
in
th
e
ex
p
er
i
m
en
t
is
p
r
esen
ted
i
n
T
ab
le
2
.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
2
5
2
-
8938
I
n
t J
A
r
ti
f
I
n
tell
,
Vo
l.
9
,
No
.
3
,
Sep
te
m
b
er
20
20
:
3
8
7
–
3
93
390
T
ab
le
2
.
T
o
tal
n
u
m
b
er
o
f
co
llected
v
o
ice
d
ata
Ty
p
e
s
D
a
t
a
Go
2
,
3
7
2
S
t
o
p
2
,
3
8
0
L
e
f
t
2
,
3
5
3
R
i
g
h
t
2
,
3
6
7
U
r
b
a
n
N
o
i
se
2
,
3
7
3
W
h
i
t
e
N
o
i
se
2
,
3
0
0
T
o
t
a
l
1
4
,
1
4
5
3
.
2
.
M
el
f
re
qu
ency
ce
ps
t
ra
l
co
ef
f
icient
s
(
M
F
CC)
MFC
C
i
s
t
h
e
tech
n
iq
u
e
u
s
ed
t
o
ex
tr
ac
t
f
ea
t
u
r
e
s
f
r
o
m
a
u
d
io
s
ig
n
al.
M
FC
C
u
s
ed
t
h
e
f
r
eq
u
en
c
y
b
an
d
lo
g
ar
ith
m
icall
y
,
w
h
ic
h
allo
w
b
etter
s
p
ee
ch
p
r
o
ce
s
s
i
n
g
[
3
1
]
.
Sig
n
al
f
ea
t
u
r
es
ar
e
f
ir
s
t
w
e
n
t
t
h
r
o
u
g
h
f
ea
t
u
r
e
ex
tr
ac
tio
n
a
n
d
later
f
ea
t
u
r
e
m
a
tch
i
n
g
.
T
h
e
p
r
o
ce
s
s
i
n
g
o
f
M
F
C
C
i
s
ap
p
lied
u
s
in
g
L
ib
r
o
s
a
li
b
r
ar
y
.
T
h
er
e
ar
e
s
i
x
s
tep
s
i
n
v
o
l
v
e
i
n
MF
C
C
f
ea
t
u
r
es
ex
tr
ac
tio
n
i.e
.
p
r
e
-
e
m
p
h
as
is
f
ilter
,
f
r
a
m
i
n
g
th
e
s
p
ee
ch
s
a
m
p
le,
w
i
n
d
o
w
i
n
g
,
f
ast
f
o
u
r
ier
tr
an
s
f
o
r
m
(
FF
T
)
,
m
el
f
ilter
b
an
k
,
an
d
lo
g
en
er
g
y
an
d
d
is
cr
ete
co
s
i
n
e
tr
an
s
f
o
r
m
(
DC
T
)
.
3
.
3
.
Da
t
a
prepa
ra
t
i
o
n
T
h
is
r
esear
ch
u
s
in
g
2
-
d
i
m
en
s
i
o
n
al
(
2
D)
C
N
Ns.
T
h
e
d
ata
is
d
ir
ec
tl
y
ta
k
en
f
r
o
m
r
a
w
MF
C
C
b
ec
au
s
e
th
e
d
atase
t
i
s
i
n
2
D
f
o
r
m
at
a
n
d
n
o
r
eq
u
ir
e
m
e
n
t
to
co
m
p
r
e
s
s
i
t.
T
h
e
s
ize
o
f
t
h
e
d
ata
h
a
s
to
b
e
i
n
th
e
s
a
m
e
f
o
r
m
at
w
h
ic
h
i
s
(
4
4
,
2
0
)
.
T
h
is
f
o
r
m
at
i
s
b
ein
g
s
et
b
ased
o
n
o
n
e
-
s
ec
o
n
d
len
g
t
h
o
f
d
ata.
I
f
t
h
e
s
ize
o
f
t
h
e
d
ata
ex
ce
ed
ed
,
th
en
t
h
e
e
x
ce
s
s
d
ata
w
il
l
b
e
cu
t
o
f
f
.
Ho
w
e
v
er
,
if
th
e
s
ize
i
s
les
s
t
h
an
t
h
e
f
o
r
m
at
s
ize
(
4
4
,
2
0
)
th
en
it
w
il
l
b
e
p
ad
w
it
h
ze
r
o
.
Fi
n
all
y
,
th
e
d
ata
is
r
ea
d
y
to
b
e
f
ee
d
in
t
h
e
2
D
C
N
Ns.
Fi
g
u
r
e
2
illu
s
tr
ates
t
h
e
d
ata
p
r
ep
ar
atio
n
p
r
o
ce
s
s
o
f
C
N
Ns
[
5
]
.
Fig
u
r
e
2
.
C
NNs d
ata
p
r
ep
ar
atio
n
3
.
4
.
SV
M
pa
r
a
m
et
er
t
un
ing
Su
p
p
o
r
t
Vec
to
r
Ma
ch
in
e
(
SV
M)
is
an
o
t
h
er
tech
n
iq
u
e
u
s
ed
to
class
i
f
y
t
h
e
v
o
ice
d
ata.
I
n
o
r
d
er
to
f
i
n
e
-
t
u
n
e
t
h
e
SVM,
t
h
er
e
ar
e
f
o
u
r
tee
n
ex
p
er
i
m
en
ts
h
av
e
b
ee
n
co
n
d
u
cted
.
T
h
e
p
ar
a
m
eter
u
s
ed
is
i
n
t
h
i
s
ex
p
er
i
m
e
n
t
i
s
a
t
y
p
e
o
f
k
er
n
el
,
C
w
h
ic
h
is
r
eg
u
lar
izatio
n
p
ar
a
m
eter
,
g
a
m
m
a
a
n
d
d
eg
r
ee
.
W
e
co
n
d
u
cted
th
r
ee
ex
p
er
i
m
e
n
ts
u
s
i
n
g
th
r
ee
d
if
f
er
en
t
k
er
n
el
f
u
n
ctio
n
s
i.e
.
p
o
ly
n
o
m
ial,
R
B
F
an
d
s
ig
m
o
id
.
T
h
e
r
esu
lt
s
h
o
w
n
t
h
at
th
e
p
o
l
y
n
o
m
ial
k
er
n
el
ac
h
ie
v
e
d
th
e
h
i
g
h
er
ac
c
u
r
ac
y
.
Ne
x
t,
w
e
co
n
d
u
cted
an
o
t
h
er
ex
p
er
im
en
ts
to
m
ea
s
u
r
e
th
e
ef
f
ec
t
o
f
C
co
n
t
in
u
e
w
it
h
t
h
e
ef
f
ec
t
o
f
g
a
m
m
a
an
d
t
h
e
ef
f
ec
t
o
f
d
eg
r
ee
.
W
e
ac
h
iv
ed
th
e
b
est
ac
cu
r
ac
y
r
es
u
l
t
th
at
i
s
7
2
.
3
9
%
w
i
th
k
er
n
el
t
y
p
e
p
o
ly
n
o
m
ial,
t
h
e
e
f
f
ec
t
o
f
C
i
s
1
0
,
th
e
e
f
f
ec
t
o
f
g
a
m
m
a
is
s
ca
le,
an
d
th
e
ef
f
ec
t
o
f
d
eg
r
ee
is
1
as sh
o
w
n
i
n
T
ab
le
3
.
T
ab
le
3
.
SVM
b
est m
o
d
el
K
e
r
n
e
l
C
G
a
mm
a
D
e
g
r
e
e
A
c
c
u
r
a
c
y
P
o
l
y
10
S
c
a
l
e
1
7
2
.
3
9
%
3
.
5
.
CNNs pa
ra
m
et
er
t
un
in
g
T
w
el
v
e
ex
p
er
i
m
en
t
s
w
er
e
co
n
d
u
cted
to
tu
n
e
C
NN
s
p
ar
am
e
t
er
.
Fro
m
th
e
ex
p
er
i
m
e
n
ts
,
w
e
d
escr
ib
ed
th
e
b
est
-
s
e
lecte
d
m
o
d
el
i
n
T
ab
le
4
[
6
]
.
T
ab
le
4
d
escr
ib
es th
e
la
y
er
s
,
k
er
n
el
s
ize,
s
tr
id
e,
f
ilte
r
s
,
p
ad
d
in
g
,
n
o
d
es,
b
ias,
an
d
ac
tiv
atio
n
f
o
r
th
e
C
NNs
m
o
d
el.
W
e
ac
h
iev
ed
9
5
.
3
0
%
ac
cu
r
ac
y
w
it
h
4
.
1
6
7
2
m
i
n
u
tes
f
r
o
m
t
h
e
i
m
p
le
m
en
ted
m
o
d
el.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
n
t J
A
r
ti
f
I
n
tell
I
SS
N:
2252
-
8938
C
o
mp
a
r
is
o
n
o
f CN
N
s
a
n
d
S
V
M
fo
r
vo
ice
co
n
tr
o
l wh
ee
lch
a
i
r
(
Mo
h
a
mma
d
S
h
a
h
r
u
l I
z
h
a
m
S
h
a
r
ifu
d
d
in
)
391
T
ab
le
4
.
T
h
e
b
est m
o
d
el
o
f
C
NNs
L
a
y
e
r
s
K
e
r
n
e
l
S
i
z
e
S
t
r
i
d
e
F
i
l
t
e
r
s
P
a
d
d
i
n
g
N
o
d
e
s
Bias
A
c
t
i
v
a
t
i
o
n
I
n
p
u
t
2
0
,
4
4
,
1
2
D
C
o
n
v
o
l
u
t
i
o
n
a
l
2
1
16
v
a
l
i
d
-
F
a
l
s
e
R
e
l
u
2
D
M
a
x
P
o
o
l
i
n
g
2
2
-
v
a
l
i
d
-
-
-
2
D
C
o
n
v
o
l
u
t
i
o
n
a
l
2
1
32
v
a
l
i
d
-
F
a
l
s
e
R
e
l
u
2
D
M
a
x
P
o
o
l
i
n
g
2
2
-
v
a
l
i
d
-
-
-
2
D
C
o
n
v
o
l
u
t
i
o
n
a
l
2
1
64
v
a
l
i
d
-
F
a
l
s
e
R
e
l
u
2
D
M
a
x
P
o
o
l
i
n
g
2
2
-
v
a
l
i
d
-
-
-
F
l
a
t
t
e
n
-
-
-
O
u
t
p
u
t
-
6
F
a
l
s
e
S
o
f
t
p
l
u
s
O
p
t
i
mi
z
e
r
A
d
a
m
L
o
ss
F
u
n
c
t
i
o
n
me
a
n
_
s
q
u
a
r
e
d
_
e
r
r
o
r
Ep
o
c
h
s
50
B
a
t
c
h
S
i
z
e
10
3
.
5
.
Co
m
pa
ri
s
o
n bet
w
ee
n SVM
a
nd
CNNs
B
ased
o
n
th
e
co
n
d
u
c
ted
ex
p
er
i
m
e
n
t,
C
N
Ns
ac
h
iev
ed
h
i
g
h
er
ac
cu
r
ac
y
t
h
at
i
s
9
5
.
3
0
%
w
it
h
co
m
p
ar
ed
to
SVM
t
h
at
i
s
7
2
.
3
9
%.
I
n
ter
m
o
f
ti
m
e,
SVM
to
o
k
s
h
o
r
ter
p
r
o
ce
s
s
in
g
ti
m
e
w
it
h
o
n
l
y
8
.
2
1
s
ec
o
n
d
r
ath
er
t
h
an
C
NNs
w
i
th
2
5
0
.
0
3
s
ec
o
n
d
s
to
r
u
n
.
T
h
e
ca
p
ab
ilit
y
o
f
f
ilt
er
i
n
g
th
e
n
o
i
s
e
in
t
h
e
v
o
ice
d
ata
in
th
e
C
NNs
f
ea
tu
r
e
ex
tr
ac
tio
n
la
y
er
p
r
o
d
u
ce
s
h
i
g
h
er
ac
cu
r
ac
y
co
m
p
ar
ed
to
SVM.
On
t
h
e
o
th
er
h
a
n
d
,
th
e
less
co
m
p
le
x
it
y
i
m
p
le
m
en
ta
tio
n
i
n
SV
M
g
i
v
e
s
h
o
r
ter
p
r
o
ce
s
s
in
g
ti
m
e.
T
ab
le
5
d
is
p
lay
s
th
e
ac
c
u
r
ac
y
an
d
ti
m
e
r
e
s
u
l
t
f
o
r
CN
Ns a
n
d
SVM.
T
ab
le
5
.
C
o
m
p
ar
is
o
n
o
f
r
esu
l
t
s
M
o
d
e
l
A
c
c
u
r
a
c
y
T
i
me
(
s)
S
V
M
7
2
.
3
9
%
8
.
2
1
C
N
N
s
9
5
.
3
0
%
2
5
0
.
0
3
3
.
6
.
H
a
rdwa
re
t
uning
T
o
d
em
o
n
s
tr
ate
th
e
i
m
p
le
m
en
tatio
n
o
f
t
h
e
v
o
ice
r
ec
o
g
n
i
tio
n
p
r
o
ce
s
s
,
w
e
i
n
te
g
r
ate
lo
g
ic
g
ates
in
t
h
e
Mo
to
r
Div
er
L
N2
9
8
N.
T
h
e
d
escr
ip
tio
n
o
f
t
h
e
lo
g
ic
g
ates
i
s
s
h
o
w
n
i
n
T
ab
le
6
.
T
h
e
lo
g
ic
g
ates
co
n
tr
o
l
th
e
m
o
v
e
m
e
n
t o
f
t
h
e
m
o
to
r
in
f
o
u
r
d
ir
ec
tio
n
s
,
i.e
.
g
o
,
s
to
p
,
lef
t,
an
d
r
ig
h
t [
6
]
.
T
ab
le
6
.
I
m
p
le
m
e
n
tatio
n
o
f
m
o
to
r
d
r
iv
er
M
o
t
o
r
D
r
i
v
e
r
P
i
n
L
N
2
9
8
N
M
o
t
o
r
D
i
r
e
c
t
i
o
n
I
n
1
I
n
2
I
n
3
I
n
4
0
0
0
0
S
t
o
p
0
1
0
1
Go
1
0
0
1
L
e
f
t
0
1
1
0
R
i
g
h
t
4.
CO
NCLU
SI
O
N
I
n
th
is
p
ap
er
,
w
e
p
r
o
p
o
s
e
an
i
n
telli
g
e
n
t
w
h
ee
lc
h
air
-
u
s
i
n
g
v
o
ice
r
ec
o
g
n
itio
n
.
T
h
er
e
ar
e
f
o
u
r
ty
p
es
o
f
v
o
ice
co
m
m
a
n
d
to
b
e
r
ec
o
g
n
i
ze
d
w
h
ic
h
ar
e
g
o
,
le
f
t,
r
i
g
h
t
a
n
d
s
to
p
.
T
h
e
d
ata
is
co
llec
ted
f
r
o
m
Go
o
g
le
an
d
s
o
m
e
o
f
t
h
e
m
ar
e
s
el
f
-
r
ec
o
r
d
e
d
.
Featu
r
es
f
r
o
m
th
e
d
ata
ar
e
ex
tr
ac
ted
u
s
in
g
M
FC
C
tec
h
n
i
q
u
es.
Fo
llo
w
ed
b
y
th
e
r
ec
o
g
n
itio
n
u
s
i
n
g
C
NN
s
an
d
SVM.
C
N
Ns
p
r
o
d
u
ce
d
h
ig
h
er
ac
c
u
r
ac
y
i.e
.
9
5
.
3
0
%
co
m
p
ar
ed
to
SVM
w
h
ic
h
is
o
n
l
y
7
2
.
3
9
%.
On
th
e
o
th
er
h
a
n
d
,
SVM
o
n
l
y
to
o
k
8
.
2
1
s
ec
o
n
d
s
w
h
ile
C
NN
s
to
o
k
2
5
0
.
0
3
s
ec
o
n
d
s
to
ex
ec
u
t
e.
T
h
er
ef
o
r
e,
C
NNs
p
r
o
d
u
ce
b
etter
r
esu
lt
b
ec
a
u
s
e
n
o
is
e
ar
e
f
ilter
ed
i
n
th
e
f
ea
t
u
r
e
ex
tr
ac
tio
n
la
y
er
b
ef
o
r
e
class
i
f
ied
in
t
h
e
cla
s
s
i
f
icatio
n
la
y
er
.
Ho
w
e
v
er
,
C
NN
s
to
o
k
lo
n
g
er
ti
m
e
d
u
e
to
th
e
co
m
p
le
x
it
y
o
f
t
h
e
n
et
w
o
r
k
s
an
d
th
e
le
s
s
co
m
p
le
x
it
y
i
m
p
le
m
e
n
tat
io
n
i
n
SVM
g
i
v
e
s
h
o
r
ter
p
r
o
ce
s
s
in
g
ti
m
e.
ACK
NO
WL
E
D
G
E
M
E
NT
S
T
h
e
au
th
o
r
s
w
o
u
ld
lik
e
to
th
an
k
th
e
Facu
l
t
y
o
f
C
o
m
p
u
ter
an
d
Ma
th
e
m
atica
l
Scie
n
ce
s
,
Un
i
v
er
s
it
i
T
ek
n
o
lo
g
i M
A
R
A
,
Ma
la
y
s
ia
f
o
r
th
e
s
u
p
p
o
r
t th
r
o
u
g
h
o
u
t th
i
s
r
esear
ch
.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
2
5
2
-
8938
I
n
t J
A
r
ti
f
I
n
tell
,
Vo
l.
9
,
No
.
3
,
Sep
te
m
b
er
20
20
:
3
8
7
–
3
93
392
RE
F
E
R
E
NC
E
S
[1
]
N.
A
k
tar,
I.
Ja
h
a
n
,
a
n
d
B
.
L
a
la,
“
V
o
ice
Re
c
o
g
n
it
i
o
n
b
a
se
d
In
tell
i
g
e
n
t
W
h
e
e
lch
a
ir
a
n
d
G
P
S
T
ra
c
k
in
g
S
y
ste
m
,
”
in
2
0
1
9
I
n
ter
n
a
ti
o
n
a
l
Co
n
fer
e
n
c
e
o
n
El
e
c
trica
l,
Co
m
p
u
ter
a
n
d
Co
mm
u
n
ica
ti
o
n
E
n
g
i
n
e
e
rin
g
(
ECCE
)
,
2
0
1
9
,
p
p
.
7
-
9.
[2
]
M
.
A
.
A
.
A
z
m
in
a
n
d
S
.
No
rd
in
,
“
V
o
ice
C
o
n
tr
o
l
F
o
r
W
h
e
e
lch
a
ir
M
o
v
e
m
e
n
t:
A
S
im
u
latio
n
,
”
Un
iv
e
rsiti
T
e
k
n
o
lo
g
i
M
A
R
A
,
2
0
1
6
.
[3
]
S
.
R.
A
v
u
tu
,
“
V
o
ice
c
o
n
tro
l
m
o
d
u
le
f
o
r
L
o
w
c
o
st
L
o
c
a
l
-
M
a
p
n
a
v
ig
a
ti
o
n
b
a
se
d
In
ti
ll
e
g
e
n
t
w
h
e
e
lch
a
ir,
”
2
0
1
7
IEE
E
7
th
I
n
t.
Ad
v
.
C
o
mp
u
t.
C
o
n
f.
,
p
p
.
6
0
9
-
6
1
3
,
2
0
1
7
.
[4
]
R.
C.
S
im
p
so
n
,
“
S
m
a
rt
w
h
e
e
lch
a
irs:
A
L
it
e
r
a
tu
re
Re
v
ie
w
,
”
J
.
Reh
a
b
il
.
Res
.
De
v
.
,
v
o
l.
4
2
,
n
o
.
4
,
p
.
4
2
3
,
2
0
0
5
.
[5
]
M. S.
S
h
a
rif
u
d
d
in
,
S
.
N
o
rd
i
n
,
a
n
d
A
.
M
o
h
d
A
li
,
“
Vo
ice
Co
n
tro
l
In
t
e
ll
ig
e
n
t
W
h
e
e
lch
a
ir
M
o
v
e
m
e
n
t
Us
in
g
CNN
s,”
in
2
0
1
9
1
st I
n
te
rn
a
ti
o
n
a
l
C
o
n
fer
e
n
c
e
o
n
Art
if
icia
l
In
telli
g
e
n
c
e
a
n
d
Da
ta
S
c
ien
c
e
s (
Ai
DAS
)
,
p
p
.
4
0
-
4
3
,
2
0
1
9
.
[6
]
D.
L
.
Ba
rriu
so
,
J.
P
é
re
z
-
M
a
rc
o
s,
D.
M
.
Jim
é
n
e
z
-
Bra
v
o
,
G
.
V
.
G
o
n
z
á
lez
,
a
n
d
J.
F
.
De
P
a
z
,
“
A
g
e
n
t
-
B
a
se
d
In
telli
g
e
n
t
In
terf
a
c
e
f
o
r
W
h
e
e
lch
a
ir
M
o
v
e
m
e
n
t
Co
n
tr
o
l,
”
S
e
n
so
rs
,
v
o
l.
1
8
,
p
p
.
1
-
3
1
,
2
0
1
8
.
[7
]
Z.
Ra
iy
a
n
,
“
De
si
g
n
o
f
a
n
A
rd
u
in
o
Ba
se
d
Vo
ice
-
Co
n
tr
o
ll
e
d
A
u
to
m
a
ted
W
h
e
e
lch
a
ir,
”
p
p
.
2
1
-
2
3
,
2
0
1
7
.
[8
]
C.
A
ru
n
a
,
A
.
D.
P
a
ra
m
e
s
w
a
ri,
M
.
M
a
li
n
i,
a
n
d
G
.
G
o
p
u
,
“
Vo
ic
e
Re
c
o
g
n
it
io
n
a
n
d
T
o
u
c
h
S
c
re
e
n
Co
n
tro
l
Ba
se
d
W
h
e
e
l
Ch
a
ir
f
o
r
P
a
ra
p
leg
ic
P
e
rso
n
s,”
2
0
1
4
In
t
.
Co
n
f.
Gr
e
e
n
Co
mp
u
t.
C
o
mm
u
n
.
El
e
c
tr.
E
n
g
.
,
p
p
.
1
-
5.
[9
]
S
.
Ch
a
u
h
a
n
,
a
.
S
.
A
ro
ra
,
a
n
d
A
.
Ka
u
l,
“
A
su
rv
e
y
o
f
e
m
e
rg
in
g
b
io
m
e
tri
c
m
o
d
a
li
ti
e
s,”
Pro
c
e
d
ia
C
o
mp
u
t.
S
c
i.
,
v
o
l.
2
,
p
p
.
2
1
3
-
2
1
8
,
2
0
1
0
.
[1
0
]
S.
A
li
m
a
n
d
A
.
K.
Ra
sh
id
,
“
S
o
m
e
Co
m
m
o
n
l
y
Us
e
d
S
p
e
e
c
h
F
e
a
tu
r
e
Ex
trac
ti
o
n
A
lg
o
rit
h
m
s,”
In
tec
h
Op
e
n
,
p
p
.
2
-
1
9
,
2
0
1
8
.
[1
1
]
Y.
L
e
c
u
n
,
Y.
Be
n
g
io
,
a
n
d
G
.
Hin
t
o
n
,
“
De
e
p
lea
rn
i
n
g
,
”
Na
t
.
In
t
.
W
k
ly.
J
.
S
c
i.
,
v
o
l.
5
2
1
,
p
p
.
4
3
6
-
4
4
4
,
2
0
1
5
.
[1
2
]
S
.
B.
Ja
d
h
a
v
,
V
.
R.
Ud
u
p
i,
S
.
B
.
P
a
ti
l,
a
n
d
D.
C
n
n
,
“
Co
n
v
o
l
u
ti
o
n
a
l
n
e
u
ra
l
n
e
tw
o
rk
s
f
o
r
lea
f
ima
g
e
-
b
a
se
d
p
lan
t
d
ise
a
se
c
las
si
f
ica
ti
o
n
,
”
v
o
l.
8
,
n
o
.
4
,
p
p
.
3
2
8
-
3
4
1
,
2
0
1
9
.
[1
3
]
A
.
S
.
Ra
z
a
v
ian
,
H.
A
z
izp
o
u
r,
J.
S
u
ll
iv
a
n
,
a
n
d
S
.
Ca
rlsso
n
,
“
CNN
fe
a
tu
re
s
o
ff
-
th
e
-
sh
e
lf
:
A
n
a
sto
u
n
d
i
n
g
b
a
se
li
n
e
f
o
r
re
c
o
g
n
it
io
n
,
”
in
IE
EE
Co
mp
u
ter
S
o
c
iety
Co
n
fer
e
n
c
e
o
n
Co
mp
u
t
e
r
Vi
sio
n
a
n
d
P
a
tt
e
rn
Rec
o
g
n
it
i
o
n
W
o
rk
sh
o
p
s
,
2
0
1
4
,
p
p
.
5
1
2
-
5
1
9
.
[1
4
]
S
.
Ib
ra
h
im
,
N.
A
.
Zu
lk
if
li
,
N.
S
a
b
ri,
A
.
A
.
S
h
a
ri,
a
n
d
M
.
R.
M
o
h
d
,
“
Rice
g
ra
in
c
l
a
ss
i
f
ica
ti
o
n
u
sin
g
m
u
lt
i
-
c
las
s
su
p
p
o
rt
v
e
c
to
r
m
a
c
h
in
e
(
S
V
M
),
”
v
o
l.
8
,
n
o
.
3
,
p
p
.
2
1
5
-
2
2
0
,
2
0
1
9
.
[1
5
]
N.
A
.
M
u
h
a
m
m
a
d
,
A
.
A
.
Na
sir,
Z.
Ib
ra
h
im
,
a
n
d
N.
S
a
b
ri
,
“
Ev
a
lu
a
ti
o
n
o
f
CNN
,
a
lex
n
e
t
a
n
d
G
o
o
g
leN
e
t
f
o
r
f
ru
it
re
c
o
g
n
it
io
n
,
”
In
d
o
n
e
s.
J
.
El
e
c
tr.
En
g
.
Co
mp
u
t.
S
c
i.
,
v
o
l.
1
2
,
n
o
.
2
,
p
p
.
4
6
8
-
4
7
5
,
2
0
1
8
.
[1
6
]
A
.
Ka
rp
a
th
y
,
G
.
T
o
d
e
rici,
S
.
S
h
e
tt
y
,
T
.
Leu
n
g
,
R.
S
u
k
th
a
n
k
a
r,
a
n
d
F
.
F
.
L
i,
“
L
a
rg
e
-
sc
a
le
v
id
e
o
c
las
sif
ic
a
ti
o
n
w
it
h
c
o
n
v
o
lu
ti
o
n
a
l
n
e
u
ra
l
n
e
tw
o
rk
s,”
i
n
Pro
c
e
e
d
in
g
s o
f
th
e
IEE
E
Co
mp
u
ter
S
o
c
iety
Co
n
fer
e
n
c
e
o
n
Co
mp
u
ter
Vi
sio
n
a
n
d
Pa
tt
e
rn
Rec
o
g
n
it
io
n
,
2
0
1
4
,
p
p
.
1
7
2
5
-
1
7
3
2
.
[1
7
]
X
.
W
a
n
g
,
A
.
M
o
h
d
A
li
,
a
n
d
P
.
A
n
g
e
lo
v
,
“
G
e
n
d
e
r
a
n
d
A
g
e
Clas
si
f
i
c
a
ti
o
n
o
f
Hu
m
a
n
F
a
c
e
s
f
o
r
A
u
to
m
a
ti
c
De
tec
ti
o
n
o
f
A
n
o
m
a
lo
u
s Hu
m
a
n
Be
h
a
v
io
u
r,
”
in
In
ter
n
a
ti
o
n
a
l
C
o
n
fer
e
n
c
e
o
n
Cy
b
e
rn
e
ti
c
s (
CY
BCONF
2
0
1
7
)
,
2
0
1
7
,
p
p
.
1
-
6.
[1
8
]
A
.
M
.
A
li
a
n
d
P
.
A
n
g
e
lo
v
,
“
A
n
o
m
a
lo
u
s
b
e
h
a
v
io
u
r
d
e
tec
ti
o
n
b
a
se
d
o
n
h
e
tero
g
e
n
e
o
u
s
d
a
ta
a
n
d
d
a
ta
f
u
sio
n
,
”
S
o
f
t
Co
mp
u
t
.
,
n
o
.
Dh
a
r
2
0
1
3
,
2
0
1
8
.
[1
9
]
N.
A
.
A
li
,
A
.
R.
S
y
a
f
e
e
z
a
,
A
.
S
.
Ja
a
f
a
r,
M
.
K.
M
.
F
.
A
li
f
,
a
n
d
N.
A
.
A
li
,
“
A
u
ti
sm
sp
e
c
tru
m
d
iso
rd
e
r
c
las
si
f
ica
ti
o
n
o
n
e
lec
tro
e
n
c
e
p
h
a
lo
g
ra
m
sig
n
a
l
u
sin
g
d
e
e
p
lea
rn
i
n
g
a
lg
o
rit
h
m
,
”
v
o
l.
9
,
n
o
.
1
,
p
p
.
9
1
-
9
9
,
2
0
2
0
.
[2
0
]
L
.
L
e
i
a
n
d
K.
S
h
e
,
“
Id
e
n
ti
ty
V
e
c
to
r
Ex
trac
ti
o
n
b
y
P
e
rc
e
p
t
u
a
l
W
a
v
e
let
P
a
c
k
e
t
En
tro
p
y
a
n
d
Co
n
v
o
lu
ti
o
n
a
l
Ne
u
ra
l
Ne
tw
o
rk
,
”
En
tro
p
y
,
v
o
l.
2
0
,
2
0
1
8
.
[2
1
]
W
.
G
u
a
n
,
“
P
e
rf
o
rm
a
n
c
e
Op
ti
m
iza
ti
o
n
o
f
S
p
e
e
c
h
Re
c
o
g
n
it
i
o
n
S
y
ste
m
w
it
h
De
e
p
Ne
u
ra
l
Ne
tw
o
rk
M
o
d
e
l,
”
O
p
t.
M
e
m.
Ne
u
ra
l
Ne
two
rk
s,
v
o
l.
2
7
,
n
o
.
4
,
p
p
.
2
7
2
-
2
8
2
,
2
0
1
8
.
[2
2
]
H.
Ra
i
a
n
d
A
.
Ya
d
a
v
,
“
Iris
r
e
c
o
g
n
it
io
n
u
si
n
g
c
o
m
b
in
e
d
su
p
p
o
rt
v
e
c
to
r
m
a
c
h
in
e
a
n
d
Ha
m
m
in
g
d
ista
n
c
e
a
p
p
ro
a
c
h
,
”
Exp
e
rt S
y
st.
A
p
p
l
.
,
v
o
l.
4
1
,
n
o
.
2
,
p
p
.
5
8
8
-
5
9
3
,
2
0
1
4
.
[2
3
]
R.
L
i,
F
.
L
e
e
,
Y.
Ya
n
,
a
n
d
Q.
C
h
e
n
,
“
F
a
c
e
Re
c
o
g
n
it
io
n
Us
in
g
V
e
c
to
r
Qu
a
n
ti
z
a
ti
o
n
Histo
g
ra
m
a
n
d
S
u
p
p
o
rt
V
e
c
to
r
M
a
c
h
in
e
Clas
sif
ier,”
in
In
ter
n
a
ti
o
n
a
l
Co
n
fer
e
n
c
e
o
n
Arti
f
icia
l
I
n
telli
g
e
n
c
e
:
T
e
c
h
n
i
q
u
e
s
a
n
d
A
p
p
l
ica
ti
o
n
s
(
AIT
A
2
0
1
6
)
,
2
0
1
6
,
n
o
.
A
it
a
,
p
p
.
1
4
4
-
1
4
9
.
[2
4
]
F
.
L
u
o
,
C.
L
i,
a
n
d
Z.
Ca
o
,
“
A
ffe
c
ti
v
e
-
f
e
a
tu
re
-
b
a
se
d
se
n
ti
m
e
n
t
a
n
a
ly
sis
u
sin
g
S
V
M
c
las
sif
ier,”
in
2
0
1
6
I
EE
E
2
0
t
h
In
ter
n
a
t
io
n
a
l
C
o
n
fer
e
n
c
e
o
n
C
o
mp
u
ter
S
u
p
p
o
rte
d
Co
o
p
e
ra
t
ive
W
o
rk
in
De
sig
n
(
CS
CW
D)
,
2
0
1
6
,
p
p
.
2
7
6
-
2
8
1
.
[2
5
]
M
.
A
h
m
a
d
,
S
.
Af
tab
,
a
n
d
I.
A
li
,
“
S
e
n
ti
m
e
n
t
A
n
a
l
y
sis
o
f
Twe
e
ts
u
sin
g
S
V
M
,
”
in
In
ter
n
a
ti
o
n
a
l
J
o
u
rn
a
l
o
f
C
o
mp
u
ter
Ap
p
li
c
a
ti
o
n
s
,
2
0
1
7
,
p
p
.
2
5
-
2
9
.
[2
6
]
M
.
Us
h
a
a
n
d
P
.
Ka
v
it
h
a
,
“
A
n
o
m
a
ly
b
a
se
d
in
tru
sio
n
d
e
tec
ti
o
n
f
o
r
8
0
2
.
1
1
n
e
tw
o
rk
s
w
it
h
o
p
ti
m
a
l
f
e
a
t
u
re
s u
sin
g
S
VM
c
las
si
f
ier,”
W
ire
l.
Ne
two
rk
s
,
v
o
l.
2
3
,
n
o
.
8
,
p
p
.
2
4
3
1
-
2
4
4
6
,
2
0
1
7
.
[2
7
]
J.
G
u
,
L
.
Wan
g
,
H.
W
a
n
g
,
a
n
d
S
.
W
a
n
g
,
“
A
n
o
v
e
l
a
p
p
ro
a
c
h
to
in
tru
si
o
n
d
e
tec
ti
o
n
u
si
n
g
S
VM
e
n
se
m
b
le
w
it
h
f
e
a
tu
re
a
u
g
m
e
n
tatio
n
,
”
C
o
mp
u
t
.
S
e
c
u
r.
,
v
o
l.
8
6
,
p
p
.
5
3
-
6
2
,
2
0
1
9
.
[2
8
]
N.
A
.
S
.
S
e
lv
a
k
u
m
a
ri
a
n
d
V
.
R
a
d
h
a
,
“
A
V
o
ice
A
c
ti
v
it
y
De
te
c
t
o
r
u
sin
g
S
V
M
a
n
d
Na
ïv
e
Ba
y
e
s
Clas
sif
i
c
a
ti
o
n
A
l
g
o
rit
h
m
,
”
in
In
ter
n
a
ti
o
n
a
l
Co
n
f
e
re
n
c
e
o
n
S
i
g
n
a
l
Pr
o
c
e
ss
in
g
a
n
d
Co
mm
u
n
ica
ti
o
n
(
ICS
PC’1
7
)
,
2
0
1
7
.
[2
9
]
W
.
A
stu
ti
a
n
d
E
.
B.
W
.
Riy
a
n
d
w
it
a
,
“
In
telli
g
e
n
t
a
u
to
m
a
ti
c
sta
rti
n
g
e
n
g
in
e
b
a
se
d
o
n
v
o
ice
re
c
o
g
n
it
io
n
sy
ste
m
,
”
Pro
c
.
-
1
4
t
h
IEE
E
S
t
u
d
e
n
t
Co
n
f.
Res
.
De
v
.
Ad
v
.
T
e
c
h
n
o
l.
Hu
ma
n
it
.
S
CORe
D 2
0
1
6
,
p
p
.
1
-
5
,
2
0
1
7
.
[3
0
]
Ha
rv
ian
to
,
L
.
A
sh
ian
ti
,
Ju
p
it
e
r,
a
n
d
S
.
Ju
n
a
e
d
i,
“
A
n
a
l
y
sis
a
n
d
V
o
ice
Re
c
o
g
n
it
io
n
in
In
d
o
n
e
sia
L
a
n
g
u
a
g
e
u
sin
g
M
F
CC an
d
S
V
M
M
e
th
o
d
,
”
C
o
mT
e
c
h
,
v
o
l.
7
,
n
o
.
2
,
p
p
.
1
3
1
-
1
3
9
,
2
0
1
6
.
[3
1
]
S
.
T
.
S
a
n
d
C
.
L
in
g
a
m
,
“
S
p
e
a
k
e
r
b
a
se
d
L
a
n
g
u
a
g
e
In
d
e
p
e
n
d
e
n
t
Iso
late
d
S
p
e
e
c
h
Re
c
o
g
n
it
i
o
n
S
y
ste
m
,
”
in
2
0
1
5
In
ter
n
a
t
io
n
a
l
C
o
n
fer
e
n
c
e
o
n
C
o
mm
u
n
ica
t
io
n
,
I
n
fo
rm
a
t
io
n
&
Co
mp
u
ti
n
g
T
e
c
h
n
o
l
o
g
y
(
ICC
ICT
)
,
p
p
.
1
-
7
,
2
0
1
5
.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
n
t J
A
r
ti
f
I
n
tell
I
SS
N:
2252
-
8938
C
o
mp
a
r
is
o
n
o
f CN
N
s
a
n
d
S
V
M
fo
r
vo
ice
co
n
tr
o
l wh
ee
lch
a
i
r
(
Mo
h
a
mma
d
S
h
a
h
r
u
l I
z
h
a
m
S
h
a
r
ifu
d
d
in
)
393
B
I
O
G
RAP
H
I
E
S
O
F
AUTH
O
RS
M
o
h
a
m
m
a
d
S
h
a
h
ru
l
Iz
h
a
m
Bin
S
h
a
rif
u
d
d
in
re
c
e
iv
e
d
h
is
Di
p
lo
m
a
in
El
e
c
tro
n
ic
En
g
in
e
e
rin
g
(Ro
b
o
ti
c
a
n
d
A
u
to
m
a
ti
o
n
)
f
ro
m
M
A
R
A
Ja
p
a
n
In
d
u
strial
I
n
stit
u
te
a
n
d
Ba
c
h
e
lo
r
o
f
In
f
o
rm
a
ti
o
n
T
e
c
h
n
o
lo
g
y
(h
o
n
s)
In
telli
g
e
n
t
S
y
ste
m
s
En
g
in
e
e
rin
g
f
ro
m
Un
iv
e
rsi
ti
T
e
k
n
o
lo
g
i
M
A
R
A
in
2
0
1
5
a
n
d
2
0
1
9
re
sp
e
c
ti
v
e
l
y
.
Cu
rre
n
tl
y
,
h
e
is
a
f
re
e
lan
c
e
d
e
v
e
lo
p
e
r
in
m
a
c
h
in
e
lea
rn
in
g
a
n
d
in
tern
e
t
o
f
th
in
g
s (i
o
T
)
a
p
p
li
c
a
ti
o
n
s.
Dr.
S
h
a
rif
a
li
ll
a
h
No
rd
in
re
c
e
iv
e
d
h
e
r
Ba
c
h
e
lo
r
o
f
In
f
o
r
m
a
ti
o
n
T
e
c
h
n
o
lo
g
y
in
2
0
0
1
f
ro
m
Un
iv
e
rsiti
Uta
ra
M
a
lay
si
a
(UU
M
),
M
a
ste
r
o
f
S
c
ien
c
e
(In
tern
e
t
Co
m
p
u
ti
n
g
)
in
2
0
0
3
f
ro
m
Un
iv
e
rsit
y
o
f
S
u
rre
y
,
UK
,
a
n
d
P
h
D i
n
Bio
i
n
f
o
rm
a
ti
c
s
f
ro
m
Un
iv
e
rs
it
i
M
a
lay
a
(UM)
in
2
0
1
0
.
S
h
e
is
c
u
rre
n
tl
y
a
S
e
n
io
r
L
e
c
tu
re
r
a
t
t
h
e
F
a
c
u
lt
y
o
f
Co
m
p
u
ter
a
n
d
M
a
t
h
e
m
a
ti
c
a
l
S
c
ien
c
e
s,
Un
iv
e
rsiti
T
e
k
n
o
lo
g
i
M
A
RA
(UiT
M
),
S
h
a
h
A
la
m
.
S
h
e
h
a
s
b
e
e
n
a
n
a
c
a
d
e
m
i
c
ian
in
UiT
M
sin
c
e
2
0
0
9
.
S
h
e
d
e
d
ica
tes
h
e
rse
lf
to
u
n
iv
e
rsity
t
e
a
c
h
in
g
a
n
d
c
o
n
d
u
c
ti
n
g
re
se
a
rc
h
.
He
r
re
se
a
r
c
h
f
ield
s
in
c
lu
d
e
b
io
d
iv
e
rsity
in
f
o
r
m
a
ti
c
s,
k
n
o
w
led
g
e
e
n
g
in
e
e
rin
g
,
a
n
d
a
rti
f
icia
l
in
tel
li
g
e
n
c
e
.
Dr
Az
li
z
a
M
o
h
d
A
li
re
c
e
iv
e
d
b
o
th
f
irst
a
n
d
m
a
ste
r
d
e
g
re
e
f
ro
m
Un
iv
e
rsiti
Uta
ra
M
a
la
y
sia
(UU
M
)
in
Ba
c
h
e
lo
r
o
f
In
f
o
r
m
a
ti
o
n
T
e
c
h
n
o
l
o
g
y
(2
0
0
1
)
a
n
d
M
a
ste
r
o
f
S
c
ien
c
e
(In
telli
g
e
n
t
Kn
o
w
led
g
e
-
Ba
s
e
d
S
y
ste
m
(2
0
0
3
)
.
He
jo
in
e
d
Un
iv
e
rsiti
T
e
k
n
o
lo
g
i
M
A
R
A
(
UiT
M
)
a
s
a
lec
tu
re
r
in
2
0
0
4
a
n
d
re
c
e
iv
e
d
a
P
h
D
in
Co
m
p
u
ter
S
c
ien
c
e
f
ro
m
Lan
c
a
ste
r
Un
iv
e
rsit
y
,
UK
.
S
h
e
d
e
d
ica
tes
h
e
rse
lf
to
u
n
iv
e
rsit
y
t
e
a
c
h
in
g
a
n
d
c
o
n
d
u
c
ti
n
g
re
se
a
rc
h
.
Cu
rre
n
tl
y
h
e
r
re
se
a
rc
h
in
tere
st
o
n
a
n
o
m
a
l
y
d
e
tec
ti
o
n
,
d
a
ta m
in
in
g
,
m
a
c
h
in
e
lea
rn
in
g
a
n
d
k
n
o
w
led
g
e
-
b
a
se
d
sy
st
e
m
s.
Evaluation Warning : The document was created with Spire.PDF for Python.