I
nd
o
ne
s
ia
n J
o
urna
l o
f
E
lect
rica
l En
g
ineering
a
nd
Co
m
p
u
t
er
Science
Vo
l.
10
,
No
.
2
,
May
201
8
,
p
p
.
6
2
3
~6
3
0
I
SS
N:
2502
-
4752
,
DOI
: 1
0
.
1
1
5
9
1
/
i
j
ee
cs
.
v
1
0
.
i2
.
p
p
623
-
6
3
0
623
J
o
ur
na
l ho
m
ep
a
g
e
:
h
ttp
:
//ia
e
s
co
r
e.
co
m/jo
u
r
n
a
ls
/in
d
ex
.
p
h
p
/
ijeec
s
Fabrica
ting M
o
di
sh M
o
bile Appli
c
a
tion for a
Visio
n
less
Cit
i
zen
to Act
as
a Visio
n
ed Cit
i
zen
J
a
na
nee
V
*
,
Sh
a
lin
i P
,
Dhiv
y
a
S
De
p
a
rtme
n
t
o
f
Co
m
p
u
ter S
c
ien
c
e
a
n
d
E
n
g
in
e
e
rin
g
,
Ra
jala
k
sh
m
i
En
g
in
e
e
rin
g
Co
ll
e
g
e
,
Ch
e
n
n
a
i,
T
a
m
il
Na
d
u
,
In
d
ia.
Art
icle
I
nfo
AB
ST
RAC
T
A
r
ticle
his
to
r
y:
R
ec
eiv
ed
No
v
7
,
2
0
1
7
R
ev
i
s
ed
J
an
2
7
,
2
0
1
8
A
cc
ep
ted
F
eb
26
,
2
0
1
8
T
e
c
h
n
o
lo
g
y
h
a
s
it
s
w
id
e
ra
n
g
e
o
f
a
p
p
li
c
a
ti
o
n
s
in
e
v
e
r
y
f
ield
p
o
ten
t
ially
e
v
e
n
o
n
th
e
M
o
b
i
le
tec
h
n
o
l
o
g
y
.
On
e
o
f
th
e
tec
h
n
o
lo
g
ies
w
h
ich
a
c
c
o
m
p
li
sh
th
e
b
li
n
d
p
e
rso
n
in
a
n
d
ro
i
d
m
o
b
il
e
is
V
ir
tu
a
l
Re
a
li
ty
.
In
sp
it
e
o
f
v
irt
u
a
l
re
a
li
ty
is
e
n
g
a
g
e
d
to
e
ff
e
c
tu
a
te
th
e
o
p
e
ra
ti
o
n
s,
th
e
b
li
n
d
p
e
rso
n
a
tt
e
n
ti
o
n
is
o
n
e
o
f
th
e
m
o
st
i
m
p
o
rtan
t
sta
ti
stics
.
A
lt
h
o
u
g
h
m
o
b
il
e
d
e
v
ice
s
in
c
lu
d
e
o
p
e
ra
ti
v
e
f
e
a
tu
re
s
a
v
a
il
a
b
le
f
o
r
a
m
a
u
ro
ti
c
u
se
rs,
th
e
u
se
r
in
ter
f
a
c
e
o
f
th
e
g
r
e
a
ter
p
a
rt
o
f
th
e
m
o
b
il
e
a
p
p
s
is
d
e
sig
n
e
d
f
o
r
sig
h
t
e
d
p
e
o
p
le.
If
th
e
y
c
a
rr
y
o
u
t
a
n
y
m
ist
a
k
e
s
to
u
se
th
e
a
p
p
s
it
m
a
y
lea
d
to
a
w
ro
n
g
c
a
ll
.
S
o
,
o
n
e
m
a
y
th
in
k
o
f
a
t
e
c
h
n
o
l
o
g
y
th
a
t
d
im
in
ish
e
s
th
e
a
n
x
iety
o
f
a
b
li
n
d
p
e
rso
n
f
o
r
u
si
n
g
a
p
p
s.
T
h
e
p
ro
p
o
se
d
s
y
ste
m
c
o
n
v
e
rts
th
e
te
x
t
in
to
a
u
d
i
o
f
o
r
g
iv
in
g
th
e
d
irec
ti
o
n
s
to
th
e
b
li
n
d
p
e
rso
n
a
b
o
u
t
t
h
e
g
e
stu
re
s
in
f
e
rre
d
.
F
o
r
su
c
h
c
o
n
v
e
rsio
n
a
tec
h
n
i
q
u
e
c
a
ll
e
d
sp
e
e
c
h
s
y
n
th
e
siz
e
r
is
u
se
d
.
A
ss
o
rted
i
n
n
o
v
a
ti
v
e
to
o
ls
a
re
u
se
d
in
m
o
b
il
e
p
h
o
n
e
s.
Bli
n
d
p
e
o
p
le
n
e
e
d
to
c
o
n
f
id
e
in
n
o
rm
a
l
p
e
rso
n
f
o
r
c
r
e
a
ti
n
g
a
n
d
u
p
d
a
t
in
g
a
c
o
n
tac
t.
Ou
r
p
r
o
jec
t
in
c
o
r
p
o
ra
tes
t
h
e
g
e
stu
re
s
f
ro
m
th
e
b
li
n
d
p
e
o
p
le
a
n
d
c
o
n
f
irm
s
th
e
g
e
stu
re
s
th
ro
u
g
h
v
o
ice
.
G
e
stu
re
c
o
n
v
e
rsio
n
is
a
c
c
o
m
p
li
sh
e
d
b
y
u
sin
g
h
a
p
ti
c
tec
h
n
o
lo
g
y
.
A
n
a
m
a
u
ro
ti
c
p
e
rso
n
c
a
n
c
re
a
te
a
n
e
w
c
o
n
tac
t
a
n
d
th
e
y
c
a
n
in
v
o
k
e
th
e
c
o
n
tac
t
b
y
u
sin
g
th
is
v
o
ice
c
o
n
f
irma
ti
o
n
.
Du
r
in
g
c
a
ll
in
g
if
th
e
re
is
m
u
lt
ip
le
c
o
n
tac
t
li
st,
c
a
ll
e
r
se
tt
in
g
a
re
in
it
ialize
d
a
n
d
a
sk
th
e
p
re
f
e
re
n
c
e
f
ro
m
th
e
b
li
n
d
p
e
o
p
le,
p
ro
g
re
ss
th
e
v
o
ice
c
a
ll
.
K
ey
w
o
r
d
s
:
G
e
stu
re
to
sp
e
e
c
h
A
lg
o
rit
h
m
Hid
d
e
n
M
a
rk
o
v
M
o
d
e
l
T
e
x
t
to
sp
e
e
c
h
c
o
n
v
e
rsio
n
V
it
e
rb
i
A
lg
o
rit
h
m
Co
p
y
rig
h
t
©
2
0
1
8
In
stit
u
te o
f
A
d
v
a
n
c
e
d
E
n
g
i
n
e
e
rin
g
a
n
d
S
c
ien
c
e
.
Al
l
rig
h
ts re
se
rv
e
d
.
C
o
r
r
e
s
p
o
nd
ing
A
uth
o
r
:
J
an
an
ee
V,
Dep
ar
t
m
en
t o
f
C
o
m
p
u
ter
Scie
n
ce
an
d
E
n
g
i
n
ee
r
in
g
,
R
aj
alak
s
h
m
i E
n
g
i
n
ee
r
in
g
C
o
ll
eg
e,
C
h
e
n
n
ai,
T
am
il Na
d
u
,
I
n
d
ia
-
6
0
2
1
0
5
.
E
m
ail: j
an
an
ee
.
v
@
r
aj
alak
s
h
m
i.e
d
u
.
in
1.
I
NT
RO
D
UCT
I
O
N
C
ell
p
h
o
n
e
s
ar
e
a
v
ital
p
ar
t
o
f
to
d
ay
’
s
l
if
e.
Ma
n
y
o
f
u
s
w
is
h
to
m
ak
e
a
ca
ll
o
r
s
e
n
d
a
m
es
s
ag
e
at
an
y
m
o
m
e
n
t
f
r
o
m
a
n
y
p
lace
.
Fo
r
v
is
u
all
y
g
r
o
p
ed
u
s
er
s
v
o
ice
b
ased
co
n
tact
lis
ts
ar
e
p
r
o
v
id
ed
w
it
h
m
a
n
y
ce
ll
p
h
o
n
es.
T
h
e
y
ca
n
s
elec
t
co
n
tacts
th
r
o
u
g
h
v
o
ice
an
d
m
ak
e
ca
lls
w
h
e
n
r
eq
u
ir
ed
.
Sp
ee
ch
R
ec
o
g
n
itio
n
an
d
C
o
n
v
er
s
io
n
w
ill
b
e
th
e
in
tr
i
n
s
ic
p
ar
t
o
f
t
h
e
ap
p
licatio
n
.
An
d
r
o
id
ac
tu
all
y
p
r
o
v
id
es
s
u
p
p
o
r
t
f
o
r
th
o
s
e
g
r
o
u
p
s
w
h
ic
h
ar
e
r
ea
ll
y
n
o
t
n
o
ticed
b
y
m
a
n
y
.
C
o
m
i
n
g
to
d
i
f
f
er
e
n
tl
y
ab
led
p
eo
p
le,
th
e
y
f
ac
e
m
o
r
e
p
r
o
b
lem
s
t
h
a
n
t
h
e
u
s
u
al
b
ein
g
.
I
t
w
i
ll
al
w
a
y
s
b
e
an
ec
s
ta
s
y
f
o
r
t
h
e
m
to
en
j
o
y
a
s
n
o
r
m
al
b
ei
n
g
w
i
th
a
ll
s
u
c
h
f
ac
to
r
s
.
T
h
e
ap
p
licatio
n
is
f
o
cu
s
ed
o
n
d
if
f
er
en
tl
y
ab
led
s
et
o
f
h
u
m
a
n
s
,
w
h
o
m
a
y
n
o
t
b
e
in
th
e
s
itu
atio
n
o
f
u
s
i
n
g
m
o
b
ile
p
h
o
n
es.
I
n
les
s
t
h
an
a
co
u
p
le
o
f
y
ea
r
s
,
g
est
u
r
e
b
ased
i
n
ter
ac
tio
n
h
as
b
ec
o
m
e
a
s
ta
n
d
ar
d
o
n
th
e
b
e
s
t
p
ar
t o
f
th
e
m
o
b
ile
d
ev
ice
s
.
I
t
i
s
an
u
p
co
m
i
n
g
ar
ea
o
f
r
esear
c
h
s
in
ce
t
o
u
ch
d
i
s
p
la
y
s
ar
e
m
o
r
e
a
n
d
m
o
r
e
p
r
esen
t
in
o
u
r
d
ail
y
li
f
e.
T
o
u
ch
s
cr
ee
n
s
p
r
o
v
id
e
a
g
r
ea
t
f
lex
ib
ilit
y
a
n
d
a
d
ir
ec
t
ac
ce
s
s
,
b
u
t
o
n
t
h
e
o
th
er
h
an
d
,
less
ac
ce
s
s
ib
le
to
v
is
u
all
y
i
m
p
air
ed
an
d
b
li
n
d
u
s
er
s
.
T
h
e
g
o
al
o
f
o
u
r
r
es
ea
r
ch
w
o
r
k
i
s
to
p
r
o
m
o
te
lo
w
v
i
s
io
n
u
s
er
s
t
h
e
in
ter
ac
tio
n
w
i
th
d
ev
ice
t
h
at
u
s
es
th
is
k
in
d
o
f
d
is
p
la
y
s
.
So
,
th
is
p
ap
er
d
ea
ls
ab
o
u
t
d
esig
n
.
Desi
g
n
o
f
to
u
ch
b
ased
m
o
b
ile
ap
p
s
f
it
f
o
r
v
i
s
u
all
y
i
m
p
air
ed
p
eo
p
le
[
1
]
.
Hap
tic
T
ec
h
n
o
lo
g
y
tak
e
s
s
u
p
er
io
r
ity
o
f
u
s
er
’
s
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
5
0
2
-
4752
I
n
d
o
n
esia
n
J
E
lec
E
n
g
&
C
o
m
p
Sci,
Vo
l.
10
,
No
.
2
,
Ma
y
2
0
1
8
:
6
2
3
–
6
3
0
624
s
en
s
iti
v
it
y
o
f
to
u
c
h
b
y
ap
p
l
y
i
n
g
ex
to
r
tio
n
,
o
s
cil
latio
n
an
d
d
ir
ec
tio
n
.
Hap
tic
r
e
f
er
s
to
s
e
n
s
i
n
g
a
n
d
d
ir
ec
tio
n
th
r
o
u
g
h
to
u
ch
.
As
to
u
ch
s
cr
ee
n
s
h
a
v
e
b
ec
o
m
e
p
r
im
a
r
y
,
it
is
p
iv
o
tal
t
h
at
to
u
ch
s
cr
ee
n
b
ased
in
ter
f
ac
es
b
e
u
s
ab
le
b
y
p
eo
p
le
w
it
h
all
ab
ilit
ie
s
,
i
n
clu
d
in
g
b
lin
d
a
n
d
v
is
u
all
y
i
m
p
air
ed
p
eo
p
le
[
2
]
.
Un
lik
e
f
u
ll
y
ac
cr
ed
ited
o
r
s
ig
h
ted
p
eo
p
le,
b
lin
d
p
eo
p
le
ca
n
n
o
t
v
i
e
w
m
ess
a
g
e
s
d
is
p
la
y
ed
o
n
th
e
s
m
ar
t
p
h
o
n
es
o
r
ca
n
n
o
t
u
s
e
t
h
e
b
asic
o
p
er
atio
n
s
lik
e
ca
lli
n
g
,
m
e
s
s
a
g
i
n
g
,
etc.
Hen
ce
,
co
m
m
u
n
ica
tio
n
v
ia
m
o
b
ile
d
ev
ice
is
a
ch
al
len
g
e
f
o
r
b
lin
d
u
s
er
s
,
w
h
o
o
f
ten
co
n
f
r
o
n
t
s
ev
er
e
ac
ce
s
s
ib
ilit
y
is
s
u
e
s
.
T
h
e
m
ai
n
p
r
o
b
le
m
s
ar
e
d
u
e
to
t
h
e
lac
k
o
f
h
ar
d
w
ar
e
k
e
y
s
,
m
a
k
in
g
it
to
u
g
h
to
q
u
i
ck
l
y
r
ea
c
h
an
a
r
ea
o
r
ac
tiv
ate
f
u
n
ctio
n
s
.
A
to
u
ch
s
cr
ee
n
h
a
s
n
o
u
n
iq
u
e
r
ef
er
en
ce
p
o
in
ts
d
is
tin
g
u
is
h
ab
le
b
y
f
ee
l.
So
,
a
b
lin
d
u
s
er
f
ee
ls
h
ar
d
to
f
ig
u
r
e
o
u
t
w
h
er
e
h
e
is
p
o
s
itio
n
ed
e
x
ac
tl
y
o
r
to
f
in
d
a
s
p
ec
if
ic
ite
m
/
f
u
n
ctio
n
.
He
n
ce
,
w
e
d
is
a
g
r
ee
t
h
at
o
p
er
ativ
e
to
u
ch
s
cr
ee
n
i
n
ter
f
ac
e
s
ca
n
b
e
e
n
h
a
n
ce
d
g
r
ea
tl
y
,
i
f
th
e
d
esi
g
n
er
s
ca
n
r
ea
lize
h
o
w
b
lin
d
p
eo
p
le
ac
tu
all
y
u
s
e
to
u
c
h
s
cr
ee
n
s
.
Hap
tic
d
ev
ice
ac
ts
a
s
a
n
i
n
p
u
t
an
d
o
u
tp
u
t
d
ev
ice,
ca
p
t
u
r
in
g
u
s
er
r
ea
l
ad
m
in
i
s
tr
atio
n
s
as
a
n
in
p
u
t
an
d
f
u
r
n
is
h
i
n
g
r
ea
lis
tic
to
u
ch
s
e
n
s
atio
n
s
[
3
]
as
an
o
u
tp
u
t
ac
co
m
m
o
d
ated
w
it
h
o
n
s
cr
ee
n
ac
ti
o
n
s
.
A
s
tec
h
n
o
lo
g
y
ad
v
an
ce
s
an
d
co
m
p
u
ter
p
o
w
e
r
e
m
er
g
e
s
,
h
ap
tic
d
e
v
ices
a
n
d
p
r
o
p
er
ties
ex
p
an
d
s
a
n
d
g
et
m
o
r
e
r
ea
li
s
tic.
T
h
i
s
tech
n
o
lo
g
y
h
as
v
er
if
ied
t
h
at
i
m
p
licit
o
b
j
ec
ts
ca
n
also
b
e
to
u
ch
ed
,
f
el
t
an
d
i
n
h
ib
ited
.
T
h
is
tech
n
o
lo
g
y
m
u
s
t
b
e
m
ad
e
f
ea
s
ib
le
f
o
r
th
e
f
air
co
s
t
an
d
th
e
h
ap
tic
d
e
v
ices
m
u
s
t b
e
m
ad
e
s
m
o
o
t
h
an
d
ea
s
ier
to
u
s
e.
Hap
tic
tech
n
o
lo
g
y
[
4
]
is
ex
te
n
s
i
v
el
y
u
s
ed
i
n
g
a
m
i
n
g
,
s
u
r
g
i
ca
l
s
i
m
u
latio
n
,
m
ed
ical
tr
ain
i
n
g
,
m
ilit
ar
y
tr
ain
i
n
g
in
v
ir
t
u
al
e
n
v
ir
o
n
m
en
t,
R
o
b
o
tics
,
Vir
t
u
al
ar
t
s
an
d
d
esi
g
n
,
m
o
b
ile
d
e
v
i
ce
s
,
r
esear
ch
a
n
d
en
ter
tai
n
m
e
n
t.
Hap
tic
ap
p
licatio
n
d
ep
en
d
s
u
p
o
n
h
i
g
h
l
y
f
u
n
ct
io
n
al
h
ar
d
w
ar
e
an
d
r
eq
u
ir
es
h
u
g
e
tr
an
s
f
o
r
m
atio
n
p
o
w
er
.
Fin
all
y
,
it
is
e
n
s
u
r
ed
th
at
th
e
h
ap
tic
tec
h
n
o
lo
g
y
is
t
h
e
r
esu
lt
f
o
r
co
m
m
u
n
icati
n
g
w
i
th
t
h
e
v
ir
tu
a
l e
n
v
ir
o
n
m
en
t.
Gen
er
ate
a
m
e
s
s
a
g
e
ap
p
licatio
n
f
o
r
te
x
t
to
v
o
ice
m
o
d
i
f
icatio
n
an
d
co
n
v
er
s
el
y
v
o
ice
i
s
en
a
b
led
u
s
in
g
on
-
d
e
m
an
d
lan
g
u
ag
e
m
o
d
el
i
n
ter
p
o
s
itio
n
[
5
]
.
T
h
is
ap
p
licatio
n
r
ec
eiv
es
y
o
u
r
m
e
s
s
a
g
e
an
d
ac
k
n
o
w
led
g
es
w
it
h
v
o
ice
n
o
t
if
ica
tio
n
b
y
p
r
o
n
o
u
n
cin
g
t
h
e
s
a
m
e.
A
s
a
p
ar
t
o
f
s
e
n
d
in
g
m
e
s
s
a
g
e,
t
h
is
ap
p
licatio
n
i
s
liab
le
f
o
r
v
o
ice
to
tex
t tr
an
s
f
er
e
n
ce
w
h
ich
i
s
u
tter
ed
b
y
u
s
er
,
an
d
ag
ai
n
te
x
t t
o
v
o
ice
to
r
ev
ie
w
m
es
s
a
g
e.
T
e
x
t to
Sp
ee
ch
i
s
also
id
en
ti
f
y
i
n
g
n
e
w
o
p
er
atio
n
s
o
u
t
w
ar
d
to
t
h
e
i
n
f
ir
m
it
y
m
ar
k
et.
Fo
r
in
s
ta
n
ce
,
s
p
ee
ch
i
n
te
g
r
ati
o
n
,
co
m
b
in
ed
w
i
t
h
s
p
ee
ch
r
ea
lizatio
n
,
co
n
f
es
s
f
o
r
co
m
m
u
n
icatio
n
w
i
t
h
m
o
b
ile
d
ev
ice
s
v
ia
c
o
m
m
o
n
la
n
g
u
a
g
e
p
r
o
ce
s
s
in
g
i
n
ter
f
ac
es.
Usi
n
g
NE
W
VI
SIO
N,
ca
lls
an
d
m
es
s
ag
e
s
ca
n
b
e
m
ad
e
u
s
i
n
g
p
atter
n
d
etec
tio
n
a
n
d
th
e
s
p
o
t
o
f
th
e
u
s
er
ca
n
b
e
f
e
tch
ed
u
s
i
n
g
Glo
b
al
P
o
s
itio
n
in
g
S
y
s
te
m
tech
n
o
lo
g
y
[
6
]
.
Fu
r
th
er
m
o
r
e,
w
e
s
t
ar
t
a
tex
t
-
to
-
s
p
ee
c
h
in
ter
f
ac
e
an
d
ac
h
ie
v
e
th
r
o
u
g
h
v
ib
r
atio
n
s
to
co
m
f
o
r
t
t
h
e
u
s
a
g
e
o
f
s
m
ar
t
p
h
o
n
e
s
f
o
r
t
h
e
b
li
n
d
u
s
er
s
.
A
ls
o
,
o
t
h
er
f
u
n
ctio
n
alitie
s
li
k
e
ca
l
lin
g
,
m
ess
a
g
in
g
,
ti
m
e,
b
atter
y
lev
e
l
etc.
ar
e
m
ad
e
s
i
m
p
le
f
o
r
th
e
v
is
u
all
y
ch
a
llen
g
ed
u
s
er
s
.
A
p
p
licatio
n
li
k
e
“
Vo
ic
e
f
o
r
A
n
d
r
o
id
”,
is
i
m
p
lied
f
o
r
v
is
u
all
y
c
h
alle
n
g
ed
.
I
t
i
s
a
g
lo
b
al
tr
an
s
lato
r
f
o
r
m
ap
p
in
g
i
m
a
g
es
to
s
o
u
n
d
s
.
Oth
er
ap
p
licatio
n
s
s
u
ch
as
“M
o
b
il
e
A
cc
e
s
s
ib
il
it
y
”
h
a
v
e
c
allin
g
a
n
d
m
es
s
ag
in
g
f
ea
t
u
r
es,
b
u
t t
h
e
y
ta
k
e
v
o
ice
a
s
in
p
u
t a
n
d
ar
e
n
o
t v
er
y
p
o
te
n
t
f
o
r
I
n
d
ian
E
n
g
li
s
h
ac
ce
n
t.
T
h
e
T
ex
t
T
o
Sp
ee
ch
(
T
T
S
)
[
7
]
co
n
v
er
s
io
n
w
ith
la
n
g
u
a
g
e
t
r
an
s
latio
n
is
ac
h
ie
v
ed
f
o
r
th
e
m
o
b
ile
o
n
an
d
r
o
id
en
v
ir
o
n
m
e
n
t.
I
t i
s
a
N
atu
r
al
L
an
g
u
ag
e
P
r
o
ce
s
s
i
n
g
(
NL
P
)
m
o
d
u
le
t
h
at
a
f
f
o
r
d
s
ea
s
y
co
m
m
u
n
icatio
n
f
o
r
th
e
p
er
s
o
n
w
h
o
ca
n
n
o
t
s
p
ea
k
b
u
t
ca
n
i
n
ter
ac
t
v
er
b
all
y
.
A
l
s
o
f
o
r
th
e
p
er
s
o
n
w
h
o
ca
n
n
o
t
p
er
ce
iv
e
o
th
er
r
eg
io
n
al
la
n
g
u
a
g
es c
a
n
ch
o
o
s
e
th
e
lan
g
u
ag
e
m
a
n
u
al
l
y
b
y
u
s
i
n
g
t
h
i
s
ap
p
licat
io
n
.
T
T
S
co
n
v
er
s
io
n
w
it
h
la
n
g
u
ag
e
tr
an
s
lato
r
co
n
v
er
t
s
th
e
n
o
r
m
al
la
n
g
u
a
g
e
te
x
t
in
to
ar
tif
icia
l
f
o
r
m
u
latio
n
o
f
h
u
m
a
n
s
p
ee
ch
.
T
h
is
w
o
r
k
c
h
an
g
es
th
e
w
r
i
tte
n
te
x
t
f
o
r
m
to
a
p
h
o
n
e
m
ic
r
ep
r
esen
tatio
n
.
L
ater
,
co
n
v
er
ts
th
e
p
h
o
n
e
m
ic
r
ep
r
es
en
tatio
n
to
w
av
e
f
o
r
m
s
t
h
at
ca
n
b
e
o
u
tp
u
t
as
i
n
to
n
atio
n
s
o
u
n
d
.
NL
P
is
a
f
ield
o
f
h
u
m
a
n
-
co
m
p
u
ter
s
y
n
er
g
y
th
a
t
m
a
k
es
a
co
m
p
u
ter
to
u
n
d
er
s
tan
d
an
d
m
a
n
ip
u
la
te
h
u
m
a
n
la
n
g
u
a
g
e
te
x
t
o
r
s
p
ee
ch
.
I
n
i
tiall
y
,
g
e
t
t
h
e
i
n
p
u
t
tex
t
in
th
e
E
n
g
li
s
h
lan
g
u
ag
e.
Af
ter
g
etti
n
g
t
h
e
te
x
t,
s
ep
ar
atio
n
o
f
th
e
E
n
g
li
s
h
w
o
r
d
s
f
r
o
m
t
h
e
tex
t
is
p
er
f
o
r
m
ed
.
T
h
en
,
w
e
e
x
ec
u
te
t
h
e
l
i
b
r
ar
y
lo
o
k
u
p
to
g
et
t
h
e
p
h
o
n
e
tic
eq
u
i
v
ale
n
t
o
f
th
e
tex
t
a
n
d
ar
r
an
g
e
t
h
e
s
e
e
n
tire
p
h
o
n
etic
eq
u
i
v
ale
n
ts
i
n
a
s
er
ies
r
elev
a
n
t
to
t
h
e
te
x
t.
C
o
n
s
eq
u
e
n
tl
y
,
s
p
ee
ch
s
y
n
t
h
esi
s
is
ac
h
iev
ed
an
d
t
h
e
s
p
ee
ch
q
u
alit
y
is
r
etai
n
ed
.
T
h
e
in
ten
tio
n
in
tr
a
n
s
f
o
r
m
i
n
g
m
u
ltip
le
alg
o
r
it
h
m
s
s
u
c
h
as
1
3
p
o
in
t
f
ea
tu
r
e
ex
tr
ac
tio
n
an
d
2
3
p
o
in
t
f
ea
t
u
r
e
ex
tr
ac
tio
n
is
to
h
elp
r
ev
a
m
p
p
er
f
o
r
m
a
n
ce
.
Fo
r
p
atter
n
p
r
o
ce
s
s
in
g
t
w
o
m
aj
o
r
ac
ce
s
s
io
n
s
u
ch
as
o
n
l
in
e
an
d
o
f
f
li
n
e
p
r
o
ce
s
s
in
g
w
e
r
e
co
n
s
id
er
ed
,
o
u
t
o
f
w
h
ic
h
o
n
l
in
e
r
ef
in
i
n
g
w
as
u
s
ed
as
it
is
f
aster
t
h
a
n
o
f
f
li
n
e
p
r
o
ce
s
s
in
g
a
n
d
t
h
er
e
is
n
o
n
e
ed
to
r
ed
ee
m
t
h
e
p
atter
n
a
s
i
m
ag
e.
T
h
is
ap
p
licatio
n
t
h
r
o
u
g
h
p
atter
n
p
air
in
g
,
g
est
u
r
e
d
etec
tio
n
an
d
v
o
ice
m
ess
a
g
in
g
w
o
u
ld
m
a
k
e
d
ialin
g
an
d
m
es
s
ag
i
n
g
f
r
o
m
s
m
ar
t
p
h
o
n
es
ac
ce
s
s
ib
le
an
d
u
n
co
m
p
licated
f
o
r
v
i
s
u
a
ll
y
i
m
p
air
ed
[
8
]
.
Kh
a
n
et
al.
[
9
]
Stu
d
e
n
t
-
G
L
A
S
S
w
ea
r
ab
le
is
d
esi
g
n
ed
f
o
r
s
m
ar
t
ca
m
er
a
d
ev
ice
b
u
i
lt
w
i
th
a
p
o
w
er
f
u
l
m
icr
o
co
n
tr
o
ller
th
at
h
a
s
th
e
a
b
ilit
y
to
s
ee
w
h
at
w
e,
n
o
r
m
al
p
eo
p
le,
ar
e
s
ee
in
g
,
u
n
d
er
s
ta
n
d
s
th
e
u
s
er
v
o
ice
r
eq
u
ests
an
d
s
u
p
p
lies
t
h
e
r
elev
an
t
in
f
o
r
m
a
tio
n
u
s
i
n
g
a
u
d
ito
r
y
f
ee
d
b
ac
k
th
r
o
u
g
h
a
n
ea
r
p
h
o
n
e.
T
h
e
d
ev
ice
ai
m
s
to
i
m
p
r
o
v
e
t
h
e
q
u
alit
y
o
f
li
f
e
f
o
r
t
h
e
b
lin
d
a
n
d
v
is
u
all
y
i
m
p
a
ir
ed
p
eo
p
le
an
d
m
a
k
es
th
e
m
u
n
d
er
s
ta
n
d
t
h
eir
s
u
r
r
o
u
n
d
in
g
s
i
n
a
clea
r
w
a
y
as
clo
s
e
as to
a
n
o
r
m
al
p
er
s
o
n
at
af
f
o
r
d
ab
le
co
s
ts
.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
n
d
o
n
esia
n
J
E
lec
E
n
g
&
C
o
m
p
Sci
I
SS
N:
2502
-
4752
F
a
b
r
ica
tin
g
Mo
d
is
h
Mo
b
ile
A
p
p
lica
tio
n
fo
r
a
V
is
io
n
less
C
iti
z
en
to
A
ct
a
s
a
V
is
io
n
ed
C
itiz
e
n
(
Ja
n
a
n
ee
V
)
625
P
r
ak
ash
et
al.
[
1
0
]
im
p
le
m
en
t
ed
an
ap
p
r
o
x
im
a
tio
n
a
u
to
m
at
ed
s
tr
u
ctu
r
e,
ca
lled
Fil
ter
ed
W
all
(
FW
)
an
d
it
f
i
lter
ed
d
is
p
o
s
ed
o
f
s
u
b
s
tan
ce
f
r
o
m
O
SN
cl
ien
t
s
u
b
s
ta
n
ce
s
.
T
h
e
g
o
al
is
to
u
tili
ze
e
f
f
i
cie
n
t
c
lass
if
ica
tio
n
p
r
o
ce
d
u
r
e
to
s
ta
y
a
w
a
y
f
r
o
m
o
v
er
p
o
w
er
ed
b
y
u
n
s
u
cc
ess
f
u
l
m
es
s
ag
e
s
.
I
n
OSN
s
,
co
n
ten
t
f
i
lter
in
g
ca
n
also
b
e
ab
u
s
ed
f
o
r
a
u
n
iq
u
e,
m
o
r
e
r
ea
ctiv
e.
I
n
[
1
1
]
ex
p
lain
ed
in
te
g
r
atio
n
o
f
A
d
ap
tiv
e
W
ei
g
h
t
R
a
n
k
in
g
P
o
lic
y
(
A
W
R
P
)
w
i
th
in
telli
g
e
n
t
class
i
f
ie
r
s
(
NB
-
A
W
R
P
-
D
A
an
d
J
4
8
-
A
W
R
P
-
D
A
)
v
ia
d
y
n
a
m
ic
ag
i
n
g
f
ac
to
r
to
im
p
r
o
v
e
class
i
f
ier
s
p
o
w
er
o
f
p
r
ed
ictio
n
.
T
h
e
m
e
th
o
d
s
ar
e
u
s
ed
to
ch
o
o
s
e
th
e
b
est
s
u
b
s
et
o
f
f
ea
t
u
r
es.
I
n
[
1
2
]
in
tr
o
d
u
ce
d
a
n
e
w
f
r
a
m
e
w
o
r
k
ca
lled
Fu
zz
y
b
ased
co
n
te
x
t
u
al
r
ec
o
m
m
e
n
d
atio
n
s
y
s
te
m
f
o
r
clas
s
i
f
icatio
n
o
f
c
u
s
to
m
er
r
e
v
i
e
w
s
.
I
t
ex
tr
ac
ts
t
h
e
in
f
o
r
m
atio
n
f
r
o
m
th
e
r
ev
ie
ws
b
ased
o
n
th
e
co
n
tex
t
g
i
v
e
n
b
y
u
s
er
s
.
I
n
[
1
3
]
s
tu
d
ied
to
id
en
tify
t
h
e
b
est
class
i
f
ier
s
f
o
r
class
i
m
b
ala
n
c
ed
h
ea
lth
d
ataset
s
th
r
o
u
g
h
a
co
s
t
-
b
ased
co
m
p
ar
is
o
n
o
f
class
i
f
ier
p
er
f
o
r
m
an
c
e
.
T
h
e
u
n
eq
u
al
m
is
c
lass
if
icatio
n
co
s
ts
w
er
e
r
ep
r
esen
ted
in
a
co
s
t
m
atr
i
x
,
an
d
co
s
t
-
b
en
e
f
it.
T
h
e
f
ir
s
t
alg
o
r
it
h
m
,
t
h
e
B
au
m
–
W
elch
s
ch
e
m
e,
is
u
s
ed
to
r
e
–
ev
alu
ate
t
h
e
m
o
d
el
p
ar
a
m
eter
s
.
T
h
e
s
ec
o
n
d
alg
o
r
it
h
m
,
t
h
e
Viter
b
i
m
et
h
o
d
,
is
u
s
ed
to
esti
m
ate
t
h
e
ab
ilit
y
o
f
an
H
MM
at
d
es
cr
ib
in
g
a
p
ar
ticu
lar
o
b
s
er
v
atio
n
i
n
d
ep
en
d
en
t
o
f
th
e
d
o
m
a
in
f
o
r
w
h
ic
h
HM
Ms
ar
e
b
ein
g
ap
p
lied
.
Vis
u
aliza
tio
n
w
o
u
ld
s
ti
m
u
la
te
a
m
u
c
h
b
etter
u
n
d
er
s
ta
n
d
i
n
g
o
f
t
h
e
s
y
s
te
m
p
as
s
ag
e.
Fo
r
ex
a
m
p
le,
if
an
ad
v
en
t
u
r
er
n
o
ti
ce
s
th
at
th
e
HM
M
p
ar
am
et
er
s
h
a
v
e
alr
ea
d
y
e
n
c
o
u
n
ter
ed
,
h
e
ca
n
s
to
p
th
e
tr
a
in
i
n
g
p
r
o
ce
s
s
.
T
h
is
t
y
p
e
o
f
v
is
u
al
izatio
n
ca
n
b
e
ac
co
m
p
li
s
h
ed
b
y
e
x
p
o
s
in
g
t
h
e
p
ar
am
e
ter
m
a
tr
ices
a
s
i
m
a
g
es,
w
h
er
e
t
h
e
b
r
ig
h
t
n
es
s
o
f
an
i
m
a
g
e
lo
ca
tio
n
r
ese
m
b
les
to
th
e
r
elati
v
e
w
e
ig
h
t
o
f
an
e
n
tr
y
i
n
th
e
m
a
tr
i
x
.
As
an
o
t
h
er
ex
a
m
p
le,
w
h
e
n
a
h
u
m
a
n
s
ee
s
n
o
t
o
n
l
y
th
e
to
u
g
h
HM
M
f
o
r
d
escr
ib
in
g
an
ex
a
m
p
le
b
u
t
also
its
s
tr
en
g
th
r
elati
v
e
to
o
th
er
HM
Ms,
th
e
lab
el
ass
ig
n
ed
to
th
e
ex
a
m
p
le
i
n
r
ec
o
g
n
it
io
n
ca
n
b
e
ce
r
tif
ied
.
2.
RE
S
E
ARCH
M
E
T
H
O
D
A
cc
ep
ta
n
ce
o
f
co
n
ten
t
s
t
h
r
o
u
g
h
g
es
t
u
r
e
p
la
y
s
a
v
ital
r
o
le.
Sti
ll
v
is
u
all
y
c
h
alle
n
g
ed
p
eo
p
le
d
id
n
’
t
r
ea
ch
t
h
e
f
r
ien
d
l
y
p
r
o
cu
r
e
a
t
s
m
ar
t
m
o
b
ile
s
t
u
f
f
.
T
h
is
i
s
a
m
aj
o
r
co
n
ten
tio
n
w
h
er
e
g
es
tu
r
es
ar
e
b
ad
l
y
r
ec
o
g
n
ized
b
y
s
m
ar
t
p
h
o
n
e
s
.
T
h
er
e
lead
s
a
co
n
n
ec
t
io
n
g
ap
b
et
w
ee
n
s
m
ar
t
a
n
d
r
o
id
ap
p
an
d
th
e
u
s
er
.
Ges
tu
r
e
alo
n
e
w
ill
n
o
t
g
u
id
e
f
o
r
a
g
o
o
d
ac
ce
s
s
o
f
s
m
ar
t
p
h
o
n
es.
W
e
n
ee
d
o
n
ap
p
i
n
s
u
c
h
a
w
a
y
th
a
t
th
er
e
s
h
o
u
ld
b
e
a
r
ec
o
g
n
itio
n
s
a
y
i
n
g
th
at
t
h
e
i
n
f
o
r
m
atio
n
/
d
ata
r
eq
u
ested
b
y
t
h
e
u
s
er
is
ac
q
u
ir
ed
co
r
r
ec
tl
y
.
T
h
is
t
y
p
e
o
f
co
m
m
u
n
icatio
n
g
ap
ca
n
b
e
b
lo
wn
-
a
w
a
y
b
y
o
u
r
p
ap
er
.
Vis
u
all
y
i
m
p
air
ed
ca
n
ac
ce
s
s
o
u
r
an
d
r
o
id
k
it
as
a
n
o
r
m
al
h
u
m
a
n
b
ein
g
w
it
h
o
u
t
o
n
e’
s
h
elp
.
Fi
g
u
r
e
1
ex
p
lai
n
s
a
b
o
u
t th
e
d
esi
g
n
o
f
p
r
o
p
o
s
ed
s
y
s
te
m
i
n
d
etail.
Fig
u
r
e
1
.
Desig
n
o
f
P
r
o
p
o
s
ed
S
y
s
te
m
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
5
0
2
-
4752
I
n
d
o
n
esia
n
J
E
lec
E
n
g
&
C
o
m
p
Sci,
Vo
l.
10
,
No
.
2
,
Ma
y
2
0
1
8
:
6
2
3
–
6
3
0
626
C
ell
p
h
o
n
es
ar
e
s
m
all
co
m
f
o
r
t
ab
le
f
o
r
u
s
to
ca
r
r
y
t
h
e
m
all
a
r
o
u
n
d
.
W
ith
a
ce
ll
p
h
o
n
e,
w
e
ca
n
m
ak
e
p
h
o
n
e
ca
ll
s
,
c
h
ec
k
o
u
r
e
m
ail
s
,
s
u
r
f
t
h
e
w
eb
o
r
s
e
n
d
tex
t
m
ess
a
g
es.
R
eg
r
ettab
l
y
,
w
e
m
a
y
h
a
v
e
a
h
ar
d
ti
m
e
s
ee
in
g
o
u
r
ce
ll
p
h
o
n
e
'
s
s
cr
ee
n
in
d
ir
ec
t
s
u
n
li
g
h
t.
E
it
h
er
w
e
m
a
y
g
et
a
g
lar
e,
o
r
w
e
m
a
y
n
o
t
s
ee
an
y
th
in
g
at
all.
W
e
ca
n
p
r
ev
en
t
t
h
ese
p
r
o
b
lem
s
f
r
o
m
o
cc
u
r
r
i
n
g
t
h
e
n
e
x
t
ti
m
e
y
o
u
tak
e
y
o
u
r
ce
ll
p
h
o
n
e.
Ou
r
P
ap
er
is
f
o
cu
s
ed
o
n
t
w
o
t
y
p
es
o
f
m
o
d
e.
Ge
s
tu
r
e
m
o
d
e
an
d
Vo
ice
m
o
d
e.
T
h
is
P
ap
er
s
p
ea
k
s
o
u
t
t
h
e
co
n
v
e
n
ien
t
a
n
d
e
f
f
icien
t
u
s
a
g
e
o
f
S
m
ar
t M
o
b
ile
A
p
p
l
icatio
n
f
o
r
v
i
s
u
a
ll
y
i
m
p
air
ed
p
eo
p
le.
2
.
1
.
M
ultidi
m
en
s
io
na
l H
M
M
Gestu
r
e
s
ar
e
co
n
v
er
ted
in
to
Seq
u
en
t
ial
S
y
m
b
o
ls
.
HM
M
is
a
f
in
ite
s
ta
te,
co
n
n
ec
ted
w
i
th
Mu
lt
ip
le
T
r
an
s
itio
n
s
.
E
ac
h
s
tate
h
a
s
2
p
r
o
b
ab
ilit
y
s
ets.
O
n
e
is
d
is
cr
ete
o
u
tp
u
t
d
en
s
it
y
f
u
n
ctio
n
an
d
th
e
o
th
er
is
co
n
tin
u
o
u
s
o
u
tp
u
t
d
en
s
it
y
f
u
n
ctio
n
.
M
u
ltid
i
m
e
n
s
io
n
a
l
g
est
u
r
e
is
o
n
e
o
f
t
h
e
m
u
lti
-
p
at
h
r
e
co
g
n
itio
n
p
r
o
ce
s
s
.
B
ased
o
n
th
e
ti
m
e
s
er
ies
g
e
s
t
u
r
es
ar
e
class
if
ied
in
to
G
(
x
,
y
,
t
)
.
G:
Gestu
r
e
p
atter
n
d
r
a
w
n
i
n
an
d
ar
o
u
n
d
X
an
d
Y
ax
is
.
S
-
Set o
f
s
tate
s
A
: T
r
an
s
itio
n
P
r
o
b
ab
ilit
y
Ma
tr
ix
B
: O
u
tp
u
t P
r
o
b
ab
ilit
y
o
f
d
is
cr
ete
HM
M
|
Aij
|
: T
r
an
s
itio
n
S
tate
f
r
o
m
i a
n
d
j
|
B
j
(
x
)
|
:
x
r
ep
r
esen
t
s
C
o
n
tin
u
o
u
s
Ob
s
er
v
atio
n
|
B
j
(
OK)
|
Ok
: D
i
s
cr
ete
Ob
s
er
v
atio
n
S
y
m
b
o
l
K:
R
an
d
o
m
v
ec
to
r
“
Di
s
cr
ete
HM
M”
A
ij
>=
0
,
B
j (
OK)
>=
0
,
∀
i,
j,
k
∑
A
ij
j
=1
,
∀
i,
j
∑
B
(
Ok
)
k
=1
,
∀
k
Λ=
(
A
,
B
,
π)
Π:
I
n
itial S
tate
o
f
d
is
tr
ib
u
tio
n
2
.2
.
M
ultidi
m
en
s
io
na
l A
pp
ro
a
ch
1.
Gestu
r
e
De
f
i
n
i
n
g
2.
Dr
a
w
i
n
g
a
p
atter
n
th
at
is
ev
o
lv
ed
in
an
d
r
o
id
m
o
b
ile
b
ased
o
n
x
,
y
ax
i
s
at
ti
m
e
t.
No
te
d
o
w
n
t
h
e
p
ix
els o
f
th
e
d
r
a
w
n
p
atter
n
.
A
r
r
ay
Size
is
d
ec
id
ed
d
y
n
a
m
ica
ll
y
b
y
r
etr
iev
in
g
s
cr
ee
n
r
eso
lu
t
i
o
n
.
B
ased
o
n
th
i
s
,
b
in
ar
izatio
n
i
s
d
o
n
e
s
p
ec
if
y
in
g
x
(
t)
: n
o
o
f
r
o
w
s
,
y
(
t)
:
n
o
o
f
co
lu
m
n
s
HM
M
P
r
o
ce
d
u
r
e
A
m
u
ltid
i
m
en
s
io
n
al
HM
M
h
a
s
N
d
is
ti
n
ct
h
id
d
en
s
tates a
n
d
M
o
b
s
er
v
ab
le
s
y
m
b
o
l
s
A
: T
r
an
s
itio
n
s
tate
B
: D
is
cr
ete
o
u
tp
u
t d
is
tr
ib
u
tio
n
3.
C
o
llect
T
r
ain
in
g
Set
R
a
w
d
ata
is
p
r
e
-
p
r
o
ce
s
s
ed
b
ef
o
r
e
tr
ain
in
g
HM
M.
T
r
ain
in
g
d
ata
is
co
llected
u
s
i
n
g
ST
FT
(
S
h
o
r
t
T
er
m
Fo
u
r
ier
T
r
an
s
f
o
r
m
)
ST
FT
P
r
o
ce
s
s
:
Seg
m
e
n
t
t
h
e
s
ig
n
al
in
to
n
ar
r
o
w
ti
m
e
i
n
ter
v
als
a
n
d
tak
e
Fo
u
r
ier
T
r
an
s
f
o
r
m
f
o
r
ea
c
h
s
ig
n
al.
E
ac
h
Fo
u
r
ier
T
r
an
s
f
o
r
m
is
b
ased
o
n
th
e
T
i
m
e
Sl
ice
o
f
th
e
s
i
g
n
al,
p
r
o
v
id
in
g
ti
m
e
a
n
d
f
r
eq
u
en
c
y
i
n
f
o
r
m
atio
n
.
ST
F
T
f
u
(
t’
,
u
)
=
∫
f
(
t
)
w
(
t
−
t
′
)
.
e
−
j2
π
ut
t
dt
t’
: T
i
m
e
P
ar
a
m
eter
u
: Fr
eq
u
e
n
c
y
f
(
t)
:
A
n
al
y
ze
s
i
g
n
al
w
(
t
-
t
’
)
: W
in
d
o
w
Fu
n
ctio
n
in
g
ce
n
ter
ed
at
t to
t’
ST
FT
h
as ti
m
e
lo
ca
lizat
io
n
b
u
t n
o
f
r
eq
u
e
n
c
y
lo
ca
lizat
io
n
ST
FT
f
(
t’
)
=f
(
t
’
)
.
e
−
j2
ut
′
Gestu
r
e
R
ec
o
g
n
itio
n
:
G
*
=
ar
g
m
ax
p
(
Λ/
O)
ST
FT
Stra
teg
ies:
-
C
h
o
o
s
e
a
w
in
d
o
w
f
u
n
ctio
n
o
f
f
i
n
ite
le
n
g
t
h
-
P
lace
a
w
i
n
d
o
w
o
n
to
p
o
f
t
h
e
s
ig
n
al
at
t=0
-
T
r
u
n
ca
te
th
e
s
ig
n
al
Evaluation Warning : The document was created with Spire.PDF for Python.
I
n
d
o
n
esia
n
J
E
lec
E
n
g
&
C
o
m
p
Sci
I
SS
N:
2502
-
4752
F
a
b
r
ica
tin
g
Mo
d
is
h
Mo
b
ile
A
p
p
lica
tio
n
fo
r
a
V
is
io
n
less
C
iti
z
en
to
A
ct
a
s
a
V
is
io
n
ed
C
itiz
e
n
(
Ja
n
a
n
ee
V
)
627
-
C
o
m
p
u
te
Fo
u
r
ier
T
r
an
s
f
o
r
m
f
o
r
th
e
tr
u
n
ca
ted
s
i
g
n
a
l
1.
T
r
ain
HM
M
w
it
h
T
r
ain
in
g
d
ata
L
i
k
eli
h
o
o
d
=
P
(
O/
Λ)
HM
M
b
ased
ap
p
r
o
ac
h
ar
e
s
tu
f
f
ed
w
it
h
tr
ain
ed
d
ata
2.
E
v
alu
a
te
g
es
tu
r
es
w
i
th
tr
ai
n
ed
d
ata
Ma
tch
i
n
g
w
i
th
tr
ai
n
ed
s
ets,
g
e
s
tu
r
es a
r
e
id
en
t
if
ied
u
s
i
n
g
Vite
r
b
i a
lg
o
r
ith
m
2
.
3
.
Vit
er
bi a
lg
o
rit
h
m
:
Viter
b
i is d
ata
an
d
m
e
m
o
r
y
i
n
t
en
s
i
v
e
p
r
o
ce
d
u
r
e
f
o
r
m
atch
in
g
th
e
Min
i
m
u
m
p
ath
cl
u
s
ter
s
.
1.
Dec
o
d
e
a
d
ata
s
eq
u
en
ce
th
at
h
as b
ee
n
en
co
d
ed
b
y
f
i
n
ite
s
ta
te
p
r
o
ce
s
s
2.
Viter
b
i
A
l
g
o
r
ith
m
i
s
o
p
ti
m
al
i
n
m
a
x
i
m
u
m
l
ik
el
ih
o
o
d
s
en
s
e
3.
Viter
b
i c
alcu
lates a
s
e
m
i b
r
u
te
f
o
r
ce
esti
m
a
te
o
f
li
k
eli
h
o
o
d
f
o
r
ea
ch
p
ath
th
r
o
u
g
h
tr
ell
is
4.
T
r
ellis
: b
ased
o
n
s
tar
tin
g
s
tate
f
o
r
all
p
o
s
s
ib
le
s
eq
u
e
n
ce
ar
e
g
ath
er
ed
-
C
alc
u
lati
n
g
T
r
ellis
-
Fi
n
d
in
g
S
h
o
r
test
p
ath
-
tr
ac
e
b
ac
k
-
R
eo
r
d
er
o
u
tp
u
t b
its
L
B
G
A
l
g
o
r
ith
m
:
L
B
G
A
l
g
o
r
ith
m
s
p
lits
t
h
e
tr
ai
n
in
g
v
ec
to
r
s
in
to
2
,
4
,
……….
2
m
p
ar
titi
o
n
an
d
d
eter
m
i
n
es
t
h
e
ce
n
tr
o
id
f
o
r
ea
c
h
p
ar
titi
o
n
.
I
t is r
ef
i
n
ed
iter
ativ
e
l
y
b
y
k
-
Me
a
n
s
C
l
u
s
ter
i
n
g
.
Fig
u
r
e
2
.
C
lu
s
ter
in
g
An
al
y
s
i
s
Step
s
in
L
B
G:
Step
1
: I
n
itializatio
n
: Set
L
(
n
o
o
f
P
ar
titi
o
n
s
o
r
C
lu
s
ter
s
)
=1
.
Fin
d
t
h
e
ce
n
tr
o
id
f
o
r
all
T
r
ain
in
g
d
ata
Step
2
: Sp
litt
in
g
: D
i
v
id
e
L
in
to
2
L
s
ep
ar
atio
n
Step
3
:
C
lass
if
ica
tio
n
:
C
las
s
i
f
y
th
e
s
et
o
f
T
r
ain
in
g
d
ata
Xk
in
to
o
n
e
clu
s
ter
C
i
ac
co
r
d
in
g
to
th
e
n
ea
r
est
n
eig
h
b
o
r
r
u
le
Step
4
: Co
d
eb
o
o
k
u
p
d
atio
n
: u
p
d
ate
th
e
co
d
e
w
o
r
d
f
o
r
ev
er
y
clu
s
ter
b
y
co
m
p
u
ti
n
g
th
e
ce
n
tr
o
id
in
ea
ch
cl
u
s
ter
Step
5
:
T
er
m
in
a
tio
n
1
:
T
h
e
o
v
er
all
d
is
to
r
tio
n
D
at
ea
c
h
i
t
er
ativ
e
is
r
elate
d
to
ea
ch
v
al
u
e.
I
f
D
is
b
elo
w
th
r
es
h
o
ld
,
th
en
g
o
to
s
tep
6
,
o
t
h
er
w
is
e
g
o
to
s
tep
3
Step
6
: T
e
r
m
i
n
atio
n
2
: I
f
L
eq
u
als v
ec
to
r
q
u
an
tizatio
n
co
d
e
b
o
o
k
s
ize,
th
e
n
s
to
p
,
o
th
er
w
is
e
g
o
to
s
tep
2
2
.
4
.
Sp
ee
ch
Sy
nthe
s
izing
T
h
e
tex
t
t
h
at
is
r
etr
iev
ed
a
s
an
o
u
tp
u
t
o
f
H
MM
i
s
ta
k
en
as
an
i
n
p
u
t
o
f
s
p
ee
ch
s
y
n
t
h
e
s
iz
er
.
T
T
S
(
T
ex
t
T
o
Sp
ee
ch
s
y
n
th
e
s
izer
)
is
t
h
e
p
r
o
ce
s
s
o
f
r
ea
d
i
n
g
a
T
ex
t/W
o
r
d
alo
u
d
.
T
T
S
h
as
t
w
o
b
lo
ck
s
.
U
s
er
I
n
ter
f
ac
e
a
n
d
Data
b
ase.
U
s
e
r
in
ter
f
ac
e
co
n
v
er
ts
r
a
w
te
x
t
in
to
w
o
r
d
s
.
T
h
is
p
r
o
ce
s
s
is
ca
lled
as
P
r
e
-
p
r
o
ce
s
s
in
g
/No
r
m
aliza
t
io
n
.
T
h
e
Fro
n
t
en
d
,
th
e
n
as
s
ig
n
s
P
h
o
n
etic
tr
an
s
cr
ip
tio
n
to
ea
ch
w
o
r
d
an
d
d
iv
id
es
it
in
to
d
if
f
er
e
n
t
u
n
i
ts
li
k
e
P
h
r
ases
/cla
u
s
e
s
/s
e
n
te
n
ce
s
.
B
ac
k
e
n
d
o
f
te
n
r
ef
er
r
ed
to
a
S
y
n
t
h
esizer
co
n
v
er
t
s
th
e
S
y
m
b
o
lic
r
ep
r
esen
tatio
n
i
n
to
s
o
u
n
d
b
y
m
atc
h
in
g
t
h
e
Data
b
ase.
A
l
g
o
r
ith
m
: G
e
s
t
u
r
e
to
So
u
n
d
A
l
g
o
r
ith
m
(
GT
S)
I
n
itiall
y
co
llect
th
e
g
es
tu
r
es a
s
in
p
u
t.
C
o
n
s
id
er
A
ij
as tr
an
s
itio
n
s
tat
es f
r
o
m
i to
j
.
b
j
(
Ok
)
is
b
ased
o
n
d
is
cr
ete
o
b
s
er
v
atio
n
o
f
r
an
d
o
m
s
tates.
B
in
ar
izatio
n
: 0
&
1
ar
e
ca
lcu
l
ated
b
ased
o
n
x
(
t)
: to
w
ar
d
s
x
ax
i
s
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
5
0
2
-
4752
I
n
d
o
n
esia
n
J
E
lec
E
n
g
&
C
o
m
p
Sci,
Vo
l.
10
,
No
.
2
,
Ma
y
2
0
1
8
:
6
2
3
–
6
3
0
628
y
(
t)
: to
w
ar
d
s
y
a
x
i
s
I
f
x
(
t)
>
0
an
d
y
(
t)
>
0
th
en
,
Fo
r
(
i=1
u
p
to
C
n
)
=>
m
a
tch
g
estu
r
e
id
en
t
if
y
C
i
C
n
:
C
en
tr
o
id
o
f
n
tr
ain
i
n
g
s
et
T
r
ain
in
g
s
et
: Co
m
p
ar
is
o
n
d
o
n
e
in
co
d
eb
o
o
k
d
atab
ase
C
h
o
o
s
e
a
w
i
n
d
o
w
f
u
n
ctio
n
o
f
f
i
n
ite
le
n
g
t
h
P
lace
a
w
i
n
d
o
w
o
n
to
p
o
f
th
e
s
ig
n
al
at
t=0
T
h
e
n
tr
u
n
ca
te
t
h
e
s
ig
n
al
C
o
m
p
u
te
Fo
u
r
ier
T
r
an
s
f
o
r
m
f
o
r
th
e
tr
u
n
ca
ted
s
i
g
n
al
T
r
ain
th
e
tr
ain
i
n
g
s
e
t
E
v
alu
a
tio
n
o
f
g
est
u
r
es
u
s
i
n
g
Viter
b
i a
lg
o
r
ith
m
is
d
o
n
e
C
alcu
late
T
r
ellis
(
d
is
tan
ce
b
etw
ee
n
id
ea
l e
n
co
d
er
in
p
u
t a
n
d
ac
tu
al
r
ec
eiv
ed
s
i
g
n
al)
Fin
d
lo
w
est
w
ei
g
h
i
n
g
p
at
h
A
d
d
/ c
o
m
p
ar
e
/ selec
t th
e
d
ec
is
io
n
b
its
R
eo
r
d
er
th
e
o
u
tp
u
t b
its
T
ex
t /
s
en
ten
ce
i
s
id
en
ti
f
ied
a
n
d
s
av
ed
in
a
te
m
p
o
r
ar
y
m
e
m
o
r
y
T
ex
t is tak
en
a
s
an
i
n
p
u
t to
T
T
S
Fro
n
t e
n
d
ac
ce
p
ts
th
e
te
x
t
w
h
ile
(
f
ile
!
=E
OF)
{r
ea
d
(
f
ile)
}
Sep
ar
ate
th
e
tex
t i
n
to
to
k
e
n
s
Div
id
e
th
e
to
k
en
s
i
n
to
p
h
r
ases
/ c
lau
s
es
Fin
all
y
,
co
n
n
ec
t to
d
atab
ase
id
en
ti
f
y
in
g
t
h
e
s
o
u
n
d
f
r
o
m
t
h
e
c
o
r
r
esp
o
n
d
in
g
p
h
r
ase
s
.
2
.
5
.
Vis
ua
lly
i
m
pa
ire
d Co
nta
ct
Sea
rc
h
First r
ec
eiv
e
t
h
e
g
e
s
t
u
r
e
m
o
d
e
ON
Dr
a
w
p
atter
n
u
s
i
n
g
GT
S a
lg
o
r
ith
m
.
T
h
en
ch
ar
ac
ter
ac
k
n
o
w
l
ed
g
e
m
e
n
t is d
o
n
e
f
o
r
(
i=0
;i<5
;i++)
in
s
cr
ee
n
,
n
u
m
b
er
i
n
g
ea
ch
r
o
w
w
it
h
co
n
tac
t n
a
m
e
E
x
a
m
p
le:
AAA
AA
B
AAC
T
h
en
s
p
ea
k
er
m
o
d
e
is
ON
T
h
is
r
ea
d
s
o
u
t th
e
f
ir
s
t 5
co
n
ta
ct
lis
t u
s
i
n
g
T
T
S sy
n
t
h
esizer
L
ater
,
g
e
s
tu
r
e
m
o
d
e
i
s
ON
I
f
n
u
m
b
er
i
n
g
i
s
les
s
th
a
n
o
r
eq
u
al
to
5
C
all
is
p
lace
d
E
ls
e
if
c
h
ar
ac
ter
is
f
o
u
n
d
t
h
en
,
Sear
ch
co
n
ti
n
u
es
Sto
p
3.
RE
SU
L
T
S
A
ND
AN
AL
Y
SI
S
Gestu
r
e
i
s
ev
o
l
v
ed
w
it
h
t
h
e
co
m
p
ar
is
o
n
o
f
m
u
ltip
le
cl
u
s
t
er
s
.
E
ac
h
ch
ar
ac
ter
i
s
m
a
tch
e
d
w
it
h
‘
N
’
clu
s
ter
s
.
C
o
n
s
id
er
,
if
w
e
r
ef
er
2
ch
ar
ac
ter
s
‘
a’
an
d
‘
d
’
w
h
ich
m
atc
h
es
w
it
h
5
clu
s
ter
an
d
h
a
s
5
tr
an
s
itio
n
s
tate
s
.
First
3
cl
u
s
ter
s
ar
e
s
a
m
e
a
n
d
th
e
la
s
t
2
cl
u
s
ter
s
d
if
f
er
as
s
h
o
w
n
i
n
f
ig
u
r
e
3
.
T
h
is
t
y
p
e
o
f
v
er
if
ica
tio
n
i
s
d
o
n
e
w
it
h
t
h
e
h
elp
o
f
Vi
ter
b
i
alg
o
r
ith
m
.
Ma
x
i
m
u
m
ch
ar
ac
ter
s
ar
e
m
atch
ed
w
i
th
co
d
eb
o
o
k
.
T
est
s
et
w
ith
2
0
ex
a
m
p
le
s
o
f
s
a
m
p
le
letter
s
h
o
w
s
o
n
l
y
5
%
o
f
ch
ar
ac
ter
m
is
m
atc
h
es.
Mis
m
atc
h
es
ca
n
b
e
r
ec
o
n
n
ec
ted
b
y
t
h
e
v
is
u
all
y
i
m
p
air
ed
b
y
r
e
-
e
n
ter
i
n
g
t
h
e
g
e
s
t
u
r
es.
I
n
f
i
g
u
r
e
4
,
c
o
n
s
id
er
an
in
s
ta
n
ce
asN
-
>n
o
.
o
f
HM
M
s
tates,
K
-
>c
o
d
e
b
o
o
k
clu
s
ter
s
.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
n
d
o
n
esia
n
J
E
lec
E
n
g
&
C
o
m
p
Sci
I
SS
N:
2502
-
4752
F
a
b
r
ica
tin
g
Mo
d
is
h
Mo
b
ile
A
p
p
lica
tio
n
fo
r
a
V
is
io
n
less
C
iti
z
en
to
A
ct
a
s
a
V
is
io
n
ed
C
itiz
e
n
(
Ja
n
a
n
ee
V
)
629
Fig
u
r
e
3
.
T
r
an
s
itio
n
State
o
f
‘
a
’
an
d
‘
d
’
Fig
u
r
e
4
.
T
r
ain
in
g
Data
Set
v
s
C
lu
s
ter
s
T
T
S
s
ea
r
ch
:
Vis
u
all
y
i
m
p
air
e
d
ca
n
g
o
to
co
n
tact
lis
t,
s
ea
r
ch
th
e
s
p
ec
i
f
ic
p
er
s
o
n
an
d
p
lace
a
ca
ll
w
it
h
o
u
t
a
n
o
r
m
al
p
er
s
o
n
h
el
p
.
Af
ter
m
o
v
in
g
to
co
n
tact
l
is
t
t
h
e
g
e
s
t
u
r
e
m
o
d
e
g
et
s
O
N,
w
h
ic
h
ta
k
es
t
h
e
ch
ar
ac
ter
as
an
i
n
p
u
t
a
n
d
s
o
u
n
d
as
ac
k
n
o
w
led
g
e
m
en
t.
Sh
o
r
tlis
ted
co
n
tacts
ar
e
n
u
m
b
er
ed
an
d
th
e
f
ir
s
t
5
co
n
tacts
ar
e
co
n
f
ir
m
ed
th
r
o
u
g
h
v
o
ice.
I
f
t
h
e
n
u
m
b
er
is
p
r
ess
ed
,
ca
ll
is
p
lace
d
to
th
at
co
r
r
es
p
o
n
d
in
g
p
er
s
o
n
.
I
f
th
e
u
s
er
ca
n
n
o
t
f
i
n
d
th
e
co
n
t
ac
t
f
r
o
m
t
h
e
s
h
o
r
tlis
t
h
e
ca
n
th
en
m
o
v
e
to
t
h
e
n
e
x
t
w
o
r
d
.
Hen
ce
,
ca
ll
ca
n
b
e
p
lace
d
b
y
a
v
i
s
u
al
l
y
i
m
p
air
ed
s
u
cc
e
s
s
f
u
ll
y
.
Fig
u
r
e
5
.
R
etr
iev
i
n
g
C
o
n
tac
ts
f
o
r
P
lacin
g
C
a
lls
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
5
0
2
-
4752
I
n
d
o
n
esia
n
J
E
lec
E
n
g
&
C
o
m
p
Sci,
Vo
l.
10
,
No
.
2
,
Ma
y
2
0
1
8
:
6
2
3
–
6
3
0
630
4.
CO
NCLU
SI
O
N
T
h
e
g
r
ea
ter
p
ar
t
o
f
th
e
e
x
is
ti
n
g
m
o
b
ile
A
p
p
s
h
av
e
b
ee
n
o
r
ig
i
n
all
y
cr
ea
ted
f
o
r
s
i
g
h
ted
p
eo
p
le.
T
h
is
p
ap
er
d
ea
ls
ar
o
u
n
d
m
o
b
ile
ap
p
s
d
esig
n
,
co
n
ce
n
tr
atin
g
m
ain
l
y
o
n
a
m
a
u
r
o
ticu
s
er
s
.
Sp
ee
ch
s
y
n
t
h
es
is
h
a
s
f
ar
a
w
a
y
b
ee
n
a
n
i
m
p
er
ati
v
ete
ch
n
o
lo
g
y
to
o
l
an
d
it
s
ap
p
licatio
n
i
n
t
h
is
f
iled
is
d
e
n
o
tin
g
a
n
d
b
o
u
n
d
les
s
.
I
t
allo
w
s
e
n
v
ir
o
n
m
e
n
tal
o
b
s
tacle
s
to
b
e
eli
m
i
n
ated
f
o
r
p
eo
p
le
w
it
h
i
m
p
air
m
en
t.
W
e
h
a
v
e
s
u
g
g
ested
a
n
ap
p
r
o
ac
h
f
o
r
d
esig
n
i
n
g
,
r
ec
o
g
n
izin
g
,
a
n
d
lear
n
in
g
h
u
m
a
n
ac
tio
n
s
b
y
a
p
p
ly
i
n
g
t
h
e
Hid
d
en
Ma
r
k
o
v
Mo
d
el.
HM
M
is
a
n
ef
f
ec
tu
al
p
ar
a
m
e
tr
ic
r
ep
r
esen
tatio
n
an
d
is
v
iab
l
e
to
ch
ar
ac
ter
ize
s
to
ch
asti
c
p
r
o
ce
s
s
e
s
.
I
n
lieu
o
f
u
s
i
n
g
g
eo
m
etr
ic
f
ea
t
u
r
es,
w
e
co
n
v
er
t th
e
g
es
tu
r
es
in
to
s
u
b
s
eq
u
e
n
t
s
y
m
b
o
ls
.
HM
M
s
ar
e
h
ir
ed
to
ex
h
ib
it
th
e
g
e
s
t
u
r
es,
an
d
th
eir
p
ar
a
m
eter
s
ar
e
s
t
u
d
ie
d
f
r
o
m
th
e
tr
ai
n
i
n
g
d
ata.
T
h
e
g
estu
r
es c
a
n
b
e
id
en
ti
f
ied
b
y
as
s
ess
i
n
g
th
e
tr
ain
ed
HM
Ms.
T
h
is
ap
p
licatio
n
th
r
o
u
g
h
p
atter
n
co
o
r
d
in
atin
g
,
g
es
tu
r
e
r
ec
o
g
n
itio
n
an
d
T
ex
t
to
s
p
ee
ch
in
te
g
r
atio
n
w
o
u
ld
m
a
k
e
d
ialin
g
an
d
m
es
s
ag
i
n
g
f
r
o
m
s
m
ar
t
p
h
o
n
es
p
o
s
s
ib
le
an
d
s
m
o
o
th
f
o
r
v
is
u
all
y
i
m
p
air
ed
th
r
o
u
g
h
v
o
ice
co
n
f
ir
m
atio
n
.
Ou
r
p
ap
e
r
w
o
r
k
d
ea
ls
w
i
th
o
n
l
y
E
n
g
lis
h
lan
g
u
ag
e
a
n
d
a
lp
h
ab
ets.
I
t
ca
n
al
s
o
b
e
u
s
ed
f
o
r
o
th
er
lan
g
u
a
g
es
w
h
ich
ca
n
b
e
i
m
p
le
m
en
ted
a
s
a
f
u
t
u
r
e
w
o
r
k
.
Als
o
,
t
h
is
p
ap
er
is
d
esi
g
n
e
d
m
ai
n
l
y
f
o
r
GT
S
co
n
v
er
s
io
n
.
T
h
e
s
a
m
e
al
g
o
r
i
th
m
an
d
co
n
ce
p
t
ca
n
b
e
u
s
ed
f
o
r
o
th
er
ap
p
licati
o
n
s
li
k
e
p
la
y
i
n
g
m
u
s
ic,
d
o
w
n
lo
ad
i
n
g
ap
p
in
a
p
la
y
s
to
r
e,
k
n
o
w
in
g
c
u
r
r
en
t ti
m
e,
etci
n
an
An
d
r
o
id
m
o
b
ile.
RE
F
E
R
E
NC
E
S
[1
]
S
id
d
h
e
sh
S
i
d
d
h
e
sh
Ba
ra
v
k
a
r,
S
id
d
h
e
sh
R.
,
M
o
h
it
R.
Bo
rd
e
,
a
n
d
M
a
h
e
n
d
ra
K.
Niv
a
n
g
u
n
e
.
"
A
n
d
ro
id
tex
t
m
e
s
sa
g
in
g
a
p
p
li
c
a
ti
o
n
f
o
r
v
isu
a
ll
y
i
m
p
a
ir
e
d
p
e
o
p
le."
IRA
C
S
T
En
g
i
n
e
e
rin
g
S
c
i
e
n
c
e
a
n
d
T
e
c
h
n
o
lo
g
y
:
a
n
In
ter
n
a
ti
o
n
a
l
J
o
u
r
n
a
l
,
v
o
l.
3
,
n
o
.
1
,
5
8
-
6
1
,
2
0
1
3
.
[2
]
S
ierra
,
Ja
v
ier
S
á
n
c
h
e
z
,
Jo
a
q
u
ín
S
e
lv
a
,
Ro
c
a
T
o
g
o
re
s,
a
n
d
Ra
y
li
g
h
t
S
o
l
u
c
io
n
e
s
T
e
c
n
o
ló
g
ica
s.
"
De
sig
n
in
g
M
o
b
il
e
Ap
p
s
fo
r
Vi
su
a
ll
y
I
mp
a
ire
d
a
n
d
Bl
in
d
Us
e
rs
Us
in
g
to
u
c
h
sc
re
e
n
b
a
se
d
mo
b
il
e
d
e
v
ice
s:
iP
h
o
n
e
/i
P
a
d
.
"
A
CHI
2
0
1
2
:T
h
e
F
if
th
In
tern
a
ti
o
n
a
l
C
o
n
f
e
re
n
c
e
o
n
A
d
v
a
n
c
e
s in
Co
m
p
u
ter
-
Hu
m
a
n
In
tera
c
ti
o
n
s
,
p
p
.
4
7
-
5
2
,
2
0
1
2
.
[3
]
J
y
o
th
i,
B.
Div
y
a
,
a
n
d
R.
V
.
Kris
h
n
a
iah
.
"
Ha
p
ti
c
T
e
c
h
n
o
lo
g
y
-
A
S
e
n
se
o
f
T
o
u
c
h
.
"
In
ter
n
a
t
io
n
a
l
J
o
u
rn
a
l
o
f
S
c
ien
c
e
a
n
d
Res
e
a
rc
h
(
IJ
S
R)
,
v
o
l.
2
,
n
o
.
9
,
p
p
.
3
8
1
-
3
8
4
,
2
0
1
3
[4
]
A
n
u
p
a
m
A
lu
r,
A
n
u
p
a
m
,
P
ra
ti
k
S
h
riv
a
sta
v
,
a
n
d
A
d
it
y
a
Ju
m
d
e
,
"
H
a
p
ti
c
tec
h
n
o
lo
g
y
:
A
c
o
m
p
re
h
e
n
siv
e
re
v
i
e
w
o
f
it
s
a
p
p
li
c
a
ti
o
n
s
a
n
d
f
u
tu
re
p
ro
sp
e
c
ts
"
In
ter
n
a
ti
o
n
a
l
J
o
u
rn
a
l
o
f
Co
m
p
u
ter
S
c
ien
c
e
a
n
d
In
fo
rm
a
ti
o
n
T
e
c
h
n
o
l
o
g
ies
,
v
o
l.
5
,
n
o
.
5
p
p
.
6
0
3
9
-
6
0
4
3
,
2
0
1
4
.
[5
]
P
a
m
p
a
tt
iw
a
r,
S
o
n
a
l
R.
,
a
n
d
A
n
il
Z.
Ch
h
a
n
g
a
n
i.
"
S
m
a
rtp
h
o
n
e
A
c
c
e
ss
ib
il
it
y
A
p
p
li
c
a
ti
o
n
f
o
r
V
isu
a
ll
y
Im
p
a
ired
.
"
In
ter
n
a
ti
o
n
a
l
J
o
u
r
n
a
l
o
f
Res
e
a
rc
h
i
n
A
d
v
e
n
t
T
e
c
h
n
o
l
o
g
y
,
2
,
n
o
.
4
,
p
p
.
3
7
7
-
3
8
0
,
2
0
1
4
.
[6
]
A
sh
ra
f
,
A
n
a
m
,
a
n
d
A
ri
f
Ra
z
a
.
"
Us
a
b
il
it
y
Iss
u
e
s
o
f
S
m
a
rt
P
h
o
n
e
A
p
p
li
c
a
ti
o
n
s:
F
o
r
V
is
u
a
ll
y
Ch
a
ll
e
n
g
e
d
P
e
o
p
le"
W
o
rld
Aca
d
e
my
o
f
S
c
ien
c
e
,
En
g
in
e
e
rin
g
a
n
d
T
e
c
h
n
o
lo
g
y
,
In
ter
n
a
ti
o
n
a
l
J
o
u
r
n
a
l
o
f
C
o
mp
u
ter
,
El
e
c
trica
l,
Au
to
m
a
ti
o
n
,
C
o
n
tro
l
a
n
d
I
n
fo
rm
a
ti
o
n
En
g
in
e
e
rin
g
,
v
o
l
.
8
,
n
o
.
5
,
p
p
.
7
6
0
-
7
6
7
,
2
0
1
4
.
[7
]
S
h
a
rm
a
,
De
v
i
k
a
,
a
n
d
Ra
n
j
u
K
a
n
w
a
r,
"
Tex
t
to
S
p
e
e
c
h
C
o
n
v
e
rsio
n
w
it
h
L
a
n
g
u
a
g
e
T
r
a
n
sla
to
r
u
n
d
e
r
A
n
d
ro
i
d
En
v
iro
n
m
e
n
t.
"
In
ter
n
a
ti
o
n
a
l
J
o
u
rn
a
l
o
f
Eme
rg
i
n
g
Res
e
a
rc
h
in
M
a
n
a
g
e
me
n
t
&
T
e
c
h
n
o
lo
g
y
,
v
o
l.
4
,
n
o
.
6
,
p
p
9
6
-
1
0
0
,
2
0
1
5
.
[8
]
Ke
tak
i
B.
T
h
a
rk
u
d
e
,
A
ish
wa
r
y
a
K.
W
a
y
a
s
e
,
P
ra
t
ik
sh
a
S
.
M
o
re
,
S
o
n
a
li
S
.
Ko
th
e
y
.
S
m
a
rt,
A
n
d
ro
id
A
p
p
li
c
a
ti
o
n
f
o
r
Bli
n
d
P
e
o
p
le
Ba
se
d
o
n
Ob
jec
t
De
tec
ti
o
n
,
In
ter
n
a
ti
o
n
a
l
J
o
u
r
n
a
l
o
f
In
n
o
v
a
ti
v
e
Res
e
a
rc
h
in
Co
mp
u
ter
a
n
d
Co
mm
u
n
ica
ti
o
n
E
n
g
i
n
e
e
rin
g
,
v
o
l.
3
,
n
o
.
1
0
,
p
p
.
6
0
1
9
-
6
0
2
0
,
2
0
1
5
.
[9
]
Kh
a
n
,
A
.
a
n
d
P
ra
k
a
sh
,
G
.
,
De
si
g
n
a
n
d
Im
p
le
m
e
n
tatio
n
o
f
S
m
a
rt
G
la
ss
w
it
h
V
o
ice
De
tec
ti
o
n
Ca
p
a
b
il
it
y
t
o
He
lp
V
isu
a
ll
y
Im
p
a
ired
P
e
o
p
le,
In
ter
n
a
ti
o
n
a
l
J
o
u
r
n
a
l
o
f
M
C
S
q
u
a
re
S
c
ien
ti
fi
c
Res
e
a
rc
h
,
v
o
l
.
9
,
n
o
.
3
,
p
p
54
-
59
,
2
0
1
7
[1
0
]
P
ra
k
a
sh
,
G
.
,
S
a
u
ra
v
,
N.,
&
Ke
th
u
,
V
.
R.
,
“
A
n
E
ff
e
c
ti
v
e
Un
d
e
sire
d
Co
n
ten
t
F
il
trati
o
n
a
n
d
P
re
d
icti
o
n
s
F
ra
m
e
w
o
rk
in
On
li
n
e
S
o
c
ial
Ne
tw
o
rk
”
,
In
ter
n
a
ti
o
n
a
l
J
o
u
rn
a
l
o
f
A
d
v
a
n
c
e
s
in
S
i
g
n
a
l
a
n
d
Ima
g
e
S
c
ien
c
e
s
,
v
o
l.
2
,
n
o
.
2
,
p
p
.
1
-
8
,
2
0
1
6
.
[1
1
]
Ola
n
re
wa
ju
,
R.
F
.
,
&
A
z
m
a
n
,
A
.
W
.
,
“
In
telli
g
e
n
t
Co
o
p
e
ra
ti
v
e
A
d
a
p
ti
v
e
W
e
ig
h
t
Ra
n
k
in
g
P
o
li
c
y
v
ia
d
y
n
a
m
ic
a
g
in
g
b
a
se
d
o
n
NB
a
n
d
J4
8
c
las
sif
iers
”
,
In
d
o
n
e
sia
n
J
o
u
rn
a
l
o
f
El
e
c
trica
l
En
g
i
n
e
e
rin
g
a
n
d
In
f
o
rm
a
ti
c
s
(
IJ
EE
I)
,
v
o
l.
5
,
n
o
.
4
,
p
p
.
3
5
7
-
3
6
5
,
2
0
1
7
.
[1
2
]
S
u
lt
h
a
n
a
,
R.
,
&
Ra
m
a
s
a
m
y
,
S
.
,
“
Co
n
tex
t
Ba
se
d
Clas
si
f
ica
ti
o
n
o
f
Re
v
ie
w
s
Us
in
g
As
so
c
iatio
n
Ru
le
M
in
in
g
,
F
u
z
z
y
L
o
g
ics
a
n
d
On
to
l
o
g
y
”
,
Bu
ll
e
ti
n
o
f
El
e
c
trica
l
En
g
i
n
e
e
rin
g
a
n
d
In
fo
r
ma
ti
c
s
,
v
o
l.
6
,
n
o
.
3
,
p
p
.
2
5
0
-
2
5
5
,
2
0
1
7
.
[1
3
]
Ra
o
,
R.
R.
,
&
M
a
k
k
it
h
a
y
a
,
K
.
,
“
L
e
a
rn
in
g
f
ro
m
a
Cla
ss
I
m
b
a
lan
c
e
d
P
u
b
li
c
He
a
l
th
Da
tas
e
t:
a
Co
st
-
b
a
se
d
Co
m
p
a
riso
n
o
f
Clas
si
f
ier
P
e
rf
o
rm
a
n
c
e
”
,
In
ter
n
a
ti
o
n
a
l
J
o
u
rn
a
l
o
f
El
e
c
trica
l
a
n
d
Co
mp
u
ter
En
g
in
e
e
rin
g
(
IJ
ECE
)
,
v
o
l.
7
,
n
o
.
4
,
p
p
.
2
2
1
5
-
2
2
2
2
,
2
0
1
7
.
Evaluation Warning : The document was created with Spire.PDF for Python.