T
E
L
K
O
M
N
I
K
A
T
elec
o
m
m
un
ica
t
io
n Co
m
pu
t
ing
E
lect
ro
nics
a
nd
Co
ntr
o
l
Vo
l.
24
,
No
.
2
,
A
p
r
il
20
26
,
p
p
.
481
~
489
I
SS
N:
1
6
9
3
-
6
9
3
0
,
DOI
: 1
0
.
1
2
9
2
8
/TE
L
KOM
NI
KA.
v
24
i
2
.
2
7
4
4
3
481
J
o
ur
na
l ho
m
ep
a
g
e
:
h
ttp
:
//jo
u
r
n
a
l.u
a
d
.
a
c.
id
/in
d
ex
.
p
h
p
/TELK
OM
N
I
K
A
Tim
e encoded
sig
na
l proces
sing
and reco
g
nition
wit
h vecto
r
qua
ntizatio
n:
a
pp
lied t
o
A
ra
bic nu
mera
ls
Abdelm
a
j
id
L
a
m
k
a
da
m
1
,
M
o
ha
m
m
ed
K
a
rim
2
1
D
e
p
a
r
t
me
n
t
o
f
M
a
t
h
e
m
a
t
i
c
s
a
n
d
C
o
mp
u
t
e
r
S
c
i
e
n
c
e
,
LI
M
A
S
La
b
o
r
a
t
o
r
y
,
F
a
c
u
l
t
y
o
f
S
c
i
e
n
c
e
s,
S
i
d
i
M
o
h
a
me
d
B
e
n
A
b
d
e
l
l
a
h
U
n
i
v
e
r
s
i
t
y
,
F
e
z
,
M
o
r
o
c
c
o
2
D
e
p
a
r
t
me
n
t
o
f
P
h
y
si
c
s,
LI
M
A
S
La
b
o
r
a
t
o
r
y
,
F
a
c
u
l
t
y
o
f
S
c
i
e
n
c
e
s
,
S
i
d
i
M
o
h
a
me
d
B
e
n
A
b
d
e
l
l
a
h
U
n
i
v
e
r
si
t
y
,
F
e
z
,
M
o
r
o
c
c
o
Art
icle
I
nfo
AB
S
T
RAC
T
A
r
ticle
his
to
r
y:
R
ec
eiv
ed
Au
g
1
,
2
0
2
5
R
ev
is
ed
Dec
23
,
2
0
2
5
Acc
ep
ted
J
an
30
,
2
0
2
6
Th
is
a
rti
c
le
p
re
se
n
ts
o
u
r
c
o
n
tri
b
u
ti
o
n
t
o
sp
e
a
k
e
r
re
c
o
g
n
it
io
n
u
si
n
g
Ara
b
i
c
n
u
m
e
ra
ls.
Th
is
re
c
o
g
n
it
io
n
is
b
a
se
d
o
n
h
y
b
ri
d
iza
ti
o
n
b
e
twe
e
n
th
e
ti
m
e
e
n
c
o
d
e
d
si
g
n
a
l
p
r
o
c
e
ss
in
g
a
n
d
re
c
o
g
n
i
ti
o
n
(
TE
S
P
AR)
tec
h
n
i
q
u
e
a
n
d
v
e
c
t
o
r
q
u
a
n
t
iza
ti
o
n
(
VQ
),
i
n
o
r
d
e
r
to
c
o
n
so
l
id
a
te
th
e
c
las
sifica
ti
o
n
ste
p
th
a
n
k
s
t
o
th
is
c
o
m
b
in
a
ti
o
n
.
T
o
se
t
u
p
a
n
e
ffe
c
ti
v
e
a
n
d
e
fficie
n
t
re
c
o
g
n
i
ti
o
n
s
y
ste
m
,
we
u
se
d
a
c
o
rp
u
s
re
c
o
rd
e
d
u
n
d
e
r
i
d
e
a
l
c
o
n
d
it
io
n
s,
m
i
n
imiz
in
g
t
h
e
d
iffere
n
c
e
s
b
e
twe
e
n
th
e
re
fe
re
n
c
e
c
o
rp
u
s
a
n
d
th
e
tes
t
c
o
rp
u
s.
We
a
lso
a
p
p
li
e
d
th
e
li
n
e
a
r
d
isc
rimin
a
n
t
a
n
a
ly
sis
(
LDA
)
tec
h
n
iq
u
e
in
o
r
d
e
r
t
o
d
isc
rimin
a
te
t
h
e
a
c
o
u
stic
v
e
c
to
rs an
d
m
in
imiz
e
th
e
re
p
re
se
n
tativ
e
sp
a
c
e
.
T
h
is h
y
b
r
id
iza
ti
o
n
i
n
d
ica
ted
a
q
u
a
n
t
ifi
a
b
le
i
n
c
re
a
se
in
th
e
sp
e
a
k
e
r
re
c
o
g
n
it
i
o
n
ra
te
with
t
h
e
t
e
n
Ara
b
ic
n
u
m
e
ra
ls (
0
–
9)
.
K
ey
w
o
r
d
s
:
Hy
b
r
id
izatio
n
L
in
ea
r
d
is
cr
im
in
an
t a
n
aly
s
is
Sp
ea
k
er
r
ec
o
g
n
itio
n
T
im
e
en
co
d
e
d
s
ig
n
al
-
p
r
o
ce
s
s
in
g
an
d
r
ec
o
g
n
itio
n
Vec
to
r
q
u
an
tizatio
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
:
Ab
d
elm
ajid
L
am
k
a
d
am
L
ab
o
r
ato
r
y
o
f
E
n
g
i
n
ee
r
in
g
,
M
o
d
elin
g
a
n
d
Sy
s
tem
s
An
aly
s
is
(
L
E
MSA)
,
Facu
lty
o
f
Scien
ce
s
Sid
i M
o
h
am
ed
B
en
Ab
d
ella
h
Un
iv
er
s
ity
Fez,
Mo
r
o
cc
o
E
m
ail: a
b
d
elm
ajid
.
lam
k
a
d
am
@
u
s
m
b
a.
ac
.
m
a
1.
I
NT
RO
D
UCT
I
O
N
Sev
er
al
m
o
d
els
an
d
tech
n
i
q
u
e
s
h
av
e
b
ee
n
d
esig
n
ed
f
o
r
s
p
ee
ch
r
ec
o
g
n
itio
n
[
1
]
,
[
2
]
.
Ho
we
v
er
,
s
p
ee
c
h
is
a
d
y
n
a
m
ic
p
h
en
o
m
en
o
n
ev
o
lv
in
g
o
v
er
tim
e;
th
ese
ex
is
tin
g
s
o
lu
tio
n
s
a
r
e
n
o
t
a
d
ap
ted
to
th
e
m
an
a
g
em
en
t
o
f
th
is
tem
p
o
r
al
asp
ec
t.
Ho
wev
er
,
th
e
tim
e
en
co
d
ed
s
ig
n
al
p
r
o
c
ess
in
g
an
d
r
ec
o
g
n
itio
n
(
T
E
SP
AR
)
tech
n
iq
u
e
h
as
b
ee
n
im
p
lem
e
n
ted
to
f
ill th
is
g
ap
an
d
i
n
teg
r
ate
th
is
tem
p
o
r
al
asp
ec
t.
T
h
e
m
ajo
r
ity
o
f
r
ec
o
g
n
itio
n
p
r
o
b
lem
s
[
3
]
,
[
4
]
s
y
s
tem
f
r
o
m
a
m
is
m
atch
b
etwe
en
th
e
tr
ain
in
g
an
d
test
co
n
d
itio
n
s
.
I
t
is
th
er
ef
o
r
e
cr
u
cial
to
r
ed
u
ce
th
ese
d
if
f
er
en
c
es
b
etwe
en
tr
ain
in
g
an
d
test
co
r
p
u
s
in
o
r
d
er
to
im
p
r
o
v
e
ex
is
tin
g
r
ec
o
g
n
itio
n
s
y
s
tem
s
.
Mo
s
t
cu
r
r
en
t
s
p
ee
ch
r
ec
o
g
n
itio
n
s
y
s
tem
s
[
5
]
,
[
6
]
u
s
u
ally
tak
e
o
n
ly
o
n
e
asp
ec
t in
to
ac
co
u
n
t,
eith
er
tem
p
o
r
al
o
r
f
r
e
q
u
en
c
y
.
Firstl
y
,
we
p
r
esen
ted
th
e
v
ec
t
o
r
q
u
an
tizatio
n
(
VQ)
[
7
]
–
[
1
0
]
an
d
T
E
SP
AR
[
1
1
]
–
[
1
6
]
tec
h
n
iq
u
es
an
d
th
eir
f
u
n
ctio
n
al
d
iag
r
a
m
s
,
as we
ll a
s
th
eir
ad
v
an
tag
es a
n
d
d
is
ad
v
an
tag
es,
wh
ich
will e
n
ab
l
e
u
s
to
ap
p
r
o
ac
h
th
e
g
lo
b
ally
co
u
p
led
r
ec
o
g
n
itio
n
s
y
s
tem
,
th
ey
tak
e
in
to
ac
co
u
n
t
th
e
ac
o
u
s
tic
ch
ar
ac
ter
is
tics
o
f
th
e
v
o
ice
s
ig
n
al
an
d
ar
e
m
o
r
e
r
esis
tan
t
to
n
o
is
e
th
at
d
eg
r
a
d
es
s
p
ee
ch
r
e
co
g
n
itio
n
q
u
ality
.
W
e
th
e
n
co
m
p
ar
ed
th
e
two
class
if
icatio
n
alg
o
r
ith
m
s
V
Q
an
d
T
E
SP
AR
,
to
ass
e
s
s
th
ese
alg
o
r
ith
m
s
’
ad
v
a
n
tag
es
an
d
d
is
ad
v
an
tag
es.
E
ac
h
o
f
th
ese
two
class
if
icatio
n
tec
h
n
iq
u
es,
V
Q
[
8
]
,
[
9
]
a
n
d
T
E
S
PAR
[
1
2
]
,
[
1
3
]
,
h
as
lim
itatio
n
s
,
s
u
ch
as
th
e
lack
o
f
a
tem
p
o
r
al
asp
ec
t f
o
r
V
Q
a
n
d
th
e
f
r
eq
u
e
n
cy
asp
ec
t f
o
r
T
E
SP
AR
.
T
h
e
b
asic
id
ea
b
eh
in
d
th
is
h
y
b
r
id
izatio
n
is
to
co
m
b
i
n
e
th
e
s
e
two
asp
ec
ts
an
d
o
th
e
r
ad
v
an
tag
es
to
im
p
r
o
v
e
r
ec
o
g
n
itio
n
r
ate.
T
o
e
n
s
u
r
e
ap
p
r
o
p
r
iate
p
er
f
o
r
m
a
n
c
e,
it
is
im
p
o
r
ta
n
t
to
h
av
e
a
co
r
p
u
s
r
ec
o
r
d
ed
u
n
d
er
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
1
6
9
3
-
6
9
3
0
TEL
KOM
NI
KA
T
elec
o
m
m
u
n
C
o
m
p
u
t E
l
C
o
n
tr
o
l
,
Vo
l.
24
,
No
.
2
,
Ap
r
il
20
26
:
4
8
1
-
4
8
9
482
o
p
tim
al
co
n
d
itio
n
s
.
An
d
wh
e
n
co
m
p
ar
in
g
th
ese
co
r
p
u
s
,
r
o
b
u
s
t,
ef
f
icien
t
an
d
ef
f
ec
tiv
e
m
eth
o
d
s
[
1
7
]
,
[
1
8
]
m
u
s
t
b
e
u
s
ed
,
wh
ic
h
ar
e
r
o
b
u
s
t
to
n
o
is
e,
s
p
ea
k
er
v
ar
iatio
n
s
an
d
with
a
g
o
o
d
r
ep
r
esen
tati
o
n
o
f
t
h
e
ac
o
u
s
tic
ch
ar
ac
ter
is
tics
.
T
o
im
p
r
o
v
e
th
e
r
ec
o
g
n
itio
n
p
r
o
ce
s
s
,
we
co
m
p
ar
e
d
th
e
e
x
is
tin
g
h
y
b
r
i
d
izatio
n
s
[
1
9
]
–
[
2
4
]
,
wh
ic
h
h
elp
ed
u
s
to
p
r
o
p
o
s
e
a
n
d
a
p
p
r
o
ac
h
a
n
ew
h
y
b
r
id
izatio
n
b
et
wee
n
th
ese
two
class
if
ier
s
V
Q
an
d
T
E
SP
AR
.
I
n
ad
d
itio
n
,
th
e
r
ep
r
esen
tatio
n
s
p
ac
e
o
f
ac
o
u
s
tic
v
ec
to
r
s
is
o
f
te
n
o
f
la
r
g
e
s
ize,
to
m
ax
im
ize
t
h
e
s
ep
ar
atio
n
o
f
th
e
class
es
th
at
ar
e
as
s
ig
n
ed
to
ea
ch
ac
o
u
s
tic
v
ec
to
r
an
d
t
o
o
b
tain
a
r
o
b
u
s
t
an
d
o
p
tim
al
r
ep
r
esen
tatio
n
,
it
is
n
ec
ess
ar
y
to
r
etain
o
n
l
y
th
e
d
is
cr
im
in
atin
g
p
a
r
am
eter
s
.
T
h
e
m
ajo
r
ity
m
eth
o
d
u
s
ed
is
lin
ea
r
d
is
cr
im
in
an
t
an
aly
s
is
(
L
DA
)
[
2
5
]
–
[
2
7
]
,
th
e
L
DA
wh
ich
m
a
k
es
it
p
o
s
s
ib
le
to
o
b
tain
d
is
cr
im
in
an
t
p
a
r
a
m
eter
s
b
y
ap
p
ly
in
g
a
lin
ea
r
tr
an
s
f
o
r
m
atio
n
o
f
th
e
in
p
u
t sp
ac
e
to
a
s
p
ac
e
o
f
r
e
d
u
ce
d
s
ize.
An
an
aly
s
is
an
d
co
m
p
a
r
is
o
n
o
f
th
e
r
esu
lts
o
b
tain
ed
ar
e
ca
r
r
i
ed
o
u
t
th
r
o
u
g
h
o
u
t
th
is
wo
r
k
,
i
n
o
r
d
e
r
to
ev
alu
ate,
v
er
if
y
an
d
r
ec
o
g
n
iz
e
o
u
r
s
p
ea
k
er
s
ac
cu
r
ately
a
n
d
with
o
u
t
f
ailu
r
e,
en
a
b
lin
g
b
etter
au
th
en
ticatio
n
u
s
in
g
ju
s
t
th
e
ten
Ar
ab
ic
n
u
m
e
r
als (
0
–
9)
.
2.
SPEE
CH
RE
CO
G
N
I
T
I
O
N
2
.
1
.
P
rinciple
T
ec
h
n
iq
u
es
th
at
m
an
ip
u
late
a
n
d
id
en
tify
th
e
s
p
ee
ch
s
ig
n
al
f
all
in
to
th
r
ee
lev
els:
(
i)
lear
n
in
g
lev
el
(
s
ig
n
al)
,
(
ii)
ex
tr
ac
tio
n
lev
el
(
v
ec
to
r
s
)
,
an
d
(
iii)
class
if
icatio
n
lev
el
(
m
o
d
els).
R
ec
o
g
n
itio
n
tech
n
iq
u
es f
ac
ilit
ate
th
e
lear
n
in
g
an
d
en
c
o
d
in
g
o
f
in
f
o
r
m
atio
n
p
r
esen
t
in
th
e
s
p
ee
ch
s
ig
n
al
f
r
o
m
th
e
d
ata
e
x
tr
ac
ted
d
u
r
in
g
th
e
an
aly
s
is
p
h
ase.
Af
ter
th
e
e
x
tr
ac
tio
n
o
f
ac
o
u
s
tic
v
ec
to
r
s
,
cl
ass
if
icatio
n
f
o
llo
ws,
wh
ich
s
e
p
ar
ates
elem
en
tar
y
s
p
ee
ch
s
eg
m
en
ts
in
to
ap
p
r
o
p
r
i
ate
class
es.
T
h
is
o
p
er
atio
n
in
v
o
lv
es
s
tatis
tical
o
r
co
n
n
ec
tio
n
i
s
t
m
o
d
elin
g
,
u
s
in
g
co
m
p
u
ter
m
eth
o
d
s
f
r
o
m
th
e
f
ield
s
o
f
s
ig
n
al
p
r
o
ce
s
s
in
g
.
T
h
en
co
m
es
th
e
r
ec
o
g
n
itio
n
p
h
ase
wh
ich
is
th
e
co
m
p
ar
is
o
n
o
f
th
e
elem
e
n
tar
y
s
eg
m
en
ts
o
f
s
p
ee
ch
p
r
e
v
io
u
s
ly
lear
n
e
d
to
m
ak
e
th
e
m
o
s
t
p
r
o
b
ab
le
d
ec
is
io
n
,
i
t
d
o
es
th
e
p
atter
n
m
atch
i
n
g
,
o
f
ten
ca
r
r
ied
o
u
t
b
y
alg
o
r
ith
m
s
s
u
ch
as:
d
y
n
am
ic
tim
e
war
p
i
n
g
(
DT
W
)
,
h
id
d
e
n
Ma
r
k
o
v
m
o
d
els
(
HM
M
)
,
V
Q
,
a
r
tific
ial
n
eu
r
al
n
etwo
r
k
(
ANN
)
,
s
u
p
p
o
r
t
v
ec
to
r
m
ac
h
i
n
e
s
(
SVM
)
,
Gau
s
s
ian
m
ix
tu
r
e
m
o
d
els
(
GM
M
)
an
d
T
E
SP
AR
.
2
.
2
.
P
r
o
blem
s
a
nd
lim
it
a
t
io
ns
R
ec
en
t
s
tu
d
ies
in
s
p
ee
ch
r
ec
o
g
n
itio
n
r
aise
th
o
r
n
y
q
u
esti
o
n
s
r
elate
d
to
i
d
en
tify
p
r
o
b
lem
s
t
h
at
r
em
ain
u
n
an
s
wer
ed
at
th
is
s
tag
e.
T
h
es
e
q
u
esti
o
n
s
ar
e
ass
o
ciate
d
with
v
ar
iab
ilit
y
ca
u
s
ed
b
y
:
−
T
h
e
s
p
ea
k
er
: a
g
e
an
d
b
eh
av
i
o
r
−
R
ec
o
r
d
in
g
co
n
d
itio
n
s
: n
o
is
e
a
n
d
eq
u
i
p
m
en
t
−
T
h
e
r
o
o
t: g
eo
g
r
ap
h
y
−
T
h
e
lan
g
u
a
g
e:
v
o
ca
b
u
lar
y
So
,
th
e
p
r
o
b
lem
with
s
p
ea
k
er
r
ec
o
g
n
itio
n
is
h
o
w
b
est
to
m
o
d
el
th
e
u
n
its
r
ep
r
esen
tin
g
th
e
s
p
ee
ch
s
ig
n
al.
T
h
er
e
ar
e
s
ev
er
al
p
o
s
s
ib
le
ty
p
es o
f
m
o
d
el
f
o
r
m
o
d
elli
n
g
th
e
p
r
o
p
e
r
ties
o
f
a
g
i
v
en
s
ig
n
al:
−
Dy
n
am
ic
m
o
d
els,
wh
ich
ch
ar
a
cter
ize
th
e
d
y
n
a
m
ic
p
r
o
p
er
ties
o
f
th
e
s
ig
n
al
−
Dete
r
m
in
is
tic
m
o
d
els,
wh
ich
e
x
p
lo
it th
e
in
tr
in
s
ic
p
r
o
p
e
r
ties
o
f
th
e
s
ig
n
al
−
Statis
t
ical
m
o
d
els,
wh
ich
ch
ar
ac
ter
ize
th
e
s
to
ch
asti
c
p
r
o
p
e
r
ties
o
f
th
e
s
ig
n
al
T
o
r
eso
lv
e
s
o
m
e
o
f
th
e
ch
alle
n
g
es
ass
o
ciate
d
with
s
p
ea
k
er
r
ec
o
g
n
itio
n
,
s
ev
er
al
s
tu
d
ies
h
a
v
e
r
ec
e
n
tly
ex
am
in
ed
th
e
ch
allen
g
es
o
f
r
ec
o
g
n
itio
n
u
n
d
er
a
v
ar
iety
o
f
co
n
d
itio
n
s
,
an
d
s
ev
er
al
h
y
b
r
id
izatio
n
s
an
d
co
m
b
in
atio
n
s
h
a
v
e
b
ee
n
im
p
le
m
en
ted
.
3.
VO
I
CE
P
R
I
N
T
CL
A
SS
I
F
I
E
RS
3
.
1
.
T
y
pes
Sp
ee
ch
r
ec
o
g
n
itio
n
alg
o
r
ith
m
s
o
r
r
ec
o
g
n
izer
s
a
r
e
also
class
if
ied
ac
co
r
d
in
g
to
th
e
s
im
p
lify
in
g
h
y
p
o
th
eses
u
n
d
er
wh
ich
th
e
y
ar
e
in
ten
d
e
d
to
o
p
e
r
ate.
T
h
er
e
ar
e
ess
en
tially
two
ty
p
es
o
f
r
ec
o
g
n
itio
n
,
d
ep
en
d
i
n
g
o
n
th
e
in
f
o
r
m
atio
n
we
wis
h
to
ex
tr
ac
t
f
r
o
m
th
e
s
p
ee
ch
s
ig
n
al:
(
i)
s
p
ea
k
er
r
ec
o
g
n
itio
n
,
to
id
e
n
tify
th
e
p
er
s
o
n
s
p
ea
k
in
g
a
n
d
(
ii)
s
p
ee
ch
r
ec
o
g
n
itio
n
,
wh
ic
h
f
o
c
u
s
es m
ain
ly
o
n
id
e
n
tify
in
g
wh
at,
is
b
ein
g
s
aid
.
T
h
er
e
ar
e
two
ty
p
es
o
f
s
p
ea
k
er
r
ec
o
g
n
itio
n
:
tex
t
-
d
e
p
en
d
e
n
t
an
d
tex
t
-
in
d
e
p
en
d
e
n
t.
T
h
e
r
e
is
also
r
ec
o
g
n
itio
n
f
o
r
a
s
in
g
le
s
p
ea
k
er
,
f
o
r
s
ev
er
al
s
p
ea
k
er
s
,
an
d
s
p
ea
k
er
-
in
d
ep
e
n
d
en
t
r
ec
o
g
n
i
tio
n
.
I
n
th
e
ca
s
e
o
f
s
p
ee
ch
r
ec
o
g
n
itio
n
,
a
d
is
tin
ctio
n
is
m
ad
e
b
etwe
en
a
s
y
s
tem
d
ed
icate
d
to
is
o
lated
wo
r
d
s
,
co
n
n
ec
ted
wo
r
d
s
an
d
a
th
ir
d
f
o
r
co
n
tin
u
o
u
s
s
p
ee
ch
.
T
h
ese
class
if
icatio
n
tech
n
iq
u
es
v
ar
y
d
ep
en
d
in
g
o
n
th
e
asp
ec
ts
o
f
th
e
s
p
ee
ch
s
ig
n
al
th
at
ar
e
co
n
s
id
er
ed
:
s
o
m
e
tak
e
i
n
to
ac
c
o
u
n
t
th
e
tem
p
o
r
al
o
r
f
r
eq
u
e
n
cy
asp
ec
t,
o
t
h
er
s
th
e
lin
ea
r
o
r
lin
ea
r
asp
ec
t,
o
th
er
s
ar
e
b
ased
o
n
th
e
s
tatis
tica
l o
r
p
r
o
b
ab
ilis
tic
asp
ec
t.
Evaluation Warning : The document was created with Spire.PDF for Python.
T
E
L
KOM
NI
KA
T
elec
o
m
m
u
n
C
o
m
p
u
t E
l Co
n
tr
o
l
Time
en
co
d
ed
s
ig
n
a
l p
r
o
ce
s
s
in
g
a
n
d
r
ec
o
g
n
itio
n
w
ith
ve
cto
r
q
u
a
n
tiz
a
tio
n
:
…
(
A
b
d
elma
ji
d
La
mka
d
a
m
)
483
3
.
2
.
Vec
t
o
r
q
ua
ntiz
a
t
io
n
T
h
is
s
ec
tio
n
p
r
esen
ts
th
e
v
ec
to
r
q
u
a
n
tizatio
n
tech
n
i
q
u
e
an
d
d
escr
ib
es
its
p
r
in
cip
le,
co
d
in
g
p
r
o
ce
s
s
,
as we
ll a
s
it
s
ad
v
an
tag
es a
n
d
li
m
itatio
n
s
.
a.
Prin
cip
le
:
t
h
e
VQ
tech
n
iq
u
e
allo
ws
ex
tr
ac
tin
g
a
d
ictio
n
ar
y
o
f
r
ep
r
esen
tativ
e
v
ec
to
r
s
(
ce
n
tr
o
id
s
)
f
r
o
m
ac
o
u
s
tic
v
ec
to
r
s
;
th
is
d
ictio
n
ar
y
r
ef
lects
th
eir
s
p
atial
d
i
s
tr
ib
u
tio
n
as
clo
s
ely
as
p
o
s
s
ib
le.
Su
ch
a
r
ep
r
esen
tatio
n
f
ac
ilit
ates
th
e
ex
p
lo
itatio
n
o
f
th
e
co
r
r
elatio
n
b
etwe
en
th
e
elem
en
ts
o
f
a
v
ec
to
r
,
th
u
s
allo
win
g
a
r
ed
u
ctio
n
in
its
d
im
en
s
io
n
ality
.
Qu
an
tizin
g
a
v
e
cto
r
am
o
u
n
ts
to
r
e
p
r
esen
tin
g
it
b
y
a
v
ec
to
r
clo
s
e
to
a
f
in
ite
d
ictio
n
ar
y
.
T
h
e
d
ictio
n
ar
y
is
o
b
tain
ed
b
y
p
ar
titi
o
n
in
g
th
e
o
r
ig
in
al
s
p
a
ce
in
to
class
es.
Ho
wev
er
,
th
e
s
ize
o
f
th
e
d
ictio
n
ar
y
p
lay
s
a
v
er
y
im
p
o
r
tan
t r
o
le
i
n
th
e
q
u
an
tizatio
n
er
r
o
r
.
b.
C
o
d
in
g
:
t
h
e
co
n
s
tr
u
ctio
n
o
f
a
d
ictio
n
ar
y
ca
n
b
e
s
u
m
m
ar
ize
d
as
f
o
llo
ws:
(
i)
t
h
e
av
er
ag
e
er
r
o
r
it
g
en
er
ates
is
d
eter
m
in
ed
f
r
o
m
an
in
itial
d
ictio
n
ar
y
.
T
h
e
alg
o
r
ith
m
s
to
p
s
if
th
e
v
alu
e
f
alls
b
elo
w
a
ce
r
tain
th
r
esh
o
l
d
an
d
(
ii)
o
t
h
er
wis
e,
ea
ch
ce
n
t
r
o
id
is
r
ep
lace
d
b
y
th
e
a
v
er
ag
e
o
f
all
v
ec
to
r
s
b
elo
n
g
in
g
to
th
e
class
r
ep
r
esen
ted
b
y
t
h
e
ce
n
t
r
o
id
,
a
n
d
th
e
n
th
e
p
r
o
ce
s
s
is
r
ep
ea
te
d
with
th
e
n
ew
d
ata
s
et.
T
h
is
alg
o
r
ith
m
lea
d
s
o
n
ly
to
a
lo
ca
l
o
p
tim
u
m
,
wh
ich
m
ak
es
th
e
ch
o
ice
o
f
in
iti
aliza
tio
n
cr
u
cial;
th
e
VQ
co
d
in
g
m
o
d
el
is
s
h
o
wn
in
Fig
u
r
e
1
.
Fig
u
r
e
1
.
VQ
co
d
in
g
m
o
d
el
c.
Ad
v
an
tag
es
:
t
h
is
tech
n
iq
u
e
o
f
f
er
s
th
e
f
o
llo
win
g
a
d
v
an
tag
es
:
b
ased
o
n
th
e
f
r
e
q
u
en
cy
d
o
m
ain
,
s
p
ec
tr
al
an
aly
s
is
,
r
ed
u
ctio
n
in
wo
r
k
i
n
g
an
d
ca
lcu
latio
n
s
p
ac
e
,
an
d
s
tatis
tical
asp
ec
t o
f
th
e
s
p
ee
ch
s
ig
n
al.
d.
Dis
ad
v
an
tag
es
(
l
im
its
)
:
t
h
is
c
lass
if
ier
s
h
o
ws
s
o
m
e
wea
k
p
o
in
ts
(
T
ab
le
1
)
s
u
ch
as:
l
ac
k
o
f
a
tem
p
o
r
al
asp
ec
t
, l
im
ited
v
o
ca
b
u
lar
y
,
a
n
d
d
is
cr
im
in
atio
n
ac
c
u
r
ac
y
p
r
o
b
lem
s
.
T
ab
le
1
.
C
o
m
p
a
r
is
o
n
cr
iter
ia
b
etwe
en
VQ
an
d
T
E
SP
AR
C
r
i
t
e
r
i
a
VQ
TESP
A
R
Te
mp
o
r
a
l
✓
F
r
e
q
u
e
n
c
y
✓
Li
n
e
a
r
✓
S
t
a
t
i
st
i
c
a
l
✓
C
o
n
n
e
c
t
i
o
n
i
s
t
✓
V
o
c
a
b
u
l
a
r
y
Li
mi
t
e
d
N
o
l
i
mi
t
e
d
S
p
e
a
k
e
r
M
o
n
o
M
u
l
t
i
3
.
3
.
T
E
SPAR
t
ec
hn
iq
ue
T
h
is
s
ec
tio
n
p
r
esen
ts
th
e
T
E
SP
AR
tech
n
iq
u
e,
in
clu
d
in
g
its
p
r
in
cip
le,
co
d
in
g
p
r
o
ce
s
s
,
as
well
as
it
s
ad
v
an
tag
es a
n
d
lim
itatio
n
s
.
a.
Prin
cip
le
:
T
E
SP
AR
is
an
ef
f
icien
t
m
eth
o
d
f
o
r
en
c
o
d
in
g
a
s
ig
n
al.
I
t
o
f
f
e
r
s
th
e
ab
ilit
y
to
p
r
o
ce
s
s
an
d
id
en
tify
tim
e
-
e
n
co
d
e
d
s
ig
n
als
an
d
to
c
h
ar
ac
ter
ize
wa
v
ef
o
r
m
s
as
a
f
u
n
ctio
n
o
f
tim
e.
T
h
e
ap
p
r
o
ac
h
is
p
ar
ticu
lar
ly
ad
v
a
n
tag
eo
u
s
in
t
h
e
ar
ea
s
o
f
v
o
ice
r
ec
o
g
n
itio
n
an
d
b
io
m
ed
ical
s
ig
n
al
an
aly
s
is
.
Un
lik
e
o
th
er
m
eth
o
d
s
th
at
f
o
cu
s
o
n
f
r
eq
u
en
cy
o
r
o
th
er
s
ig
n
al
ch
ar
ac
t
er
is
tics
,
T
E
S
PA
R
f
o
cu
s
es
o
n
h
o
w
s
ig
n
als
ch
an
g
e
o
v
er
tim
e.
T
h
e
tech
n
iq
u
e
is
b
ased
o
n
th
e
an
al
y
s
is
an
d
ex
tr
ac
tio
n
o
f
r
elev
a
n
t f
ea
tu
r
es f
r
o
m
s
p
ee
c
h
s
ig
n
als
u
s
in
g
tim
e
co
d
in
g
.
I
t
id
en
tifie
s
p
atter
n
s
an
d
s
tr
u
ctu
r
es
in
s
p
ee
ch
s
ig
n
als
th
at
ca
n
b
e
u
s
ed
f
o
r
r
ec
o
g
n
itio
n
.
An
im
p
o
r
tan
t
f
ea
t
u
r
e
o
f
T
E
SP
AR
is
its
ab
ilit
y
to
p
r
o
ce
s
s
s
ig
n
als
in
r
ea
l
tim
e
,
m
ak
in
g
it
th
e
m
eth
o
d
o
f
ch
o
ice
f
o
r
ap
p
licatio
n
s
r
eq
u
ir
in
g
a
f
ast
r
esp
o
n
s
e,
s
u
ch
as
s
p
ee
ch
r
ec
o
g
n
i
tio
n
in
v
o
ice
co
m
m
an
d
o
r
a
u
to
m
atic
tr
an
s
cr
ip
tio
n
s
y
s
tem
s
.
b.
C
o
d
in
g
:
T
E
SP
AR
an
aly
s
es
th
e
s
h
ap
e
o
f
a
s
ig
n
al
b
y
d
iv
id
i
n
g
it
i
n
to
s
eg
m
e
n
ts
o
f
eq
u
al
tim
e
an
d
th
en
ass
ig
n
in
g
s
y
m
b
o
ls
to
ea
ch
s
eg
m
en
t
ac
c
o
r
d
in
g
to
ce
r
tain
c
h
ar
ac
ter
is
tics
o
f
th
e
s
ig
n
al
in
th
at
s
eg
m
en
t.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
1
6
9
3
-
6
9
3
0
TEL
KOM
NI
KA
T
elec
o
m
m
u
n
C
o
m
p
u
t E
l
C
o
n
tr
o
l
,
Vo
l.
24
,
No
.
2
,
Ap
r
il
20
26
:
4
8
1
-
4
8
9
484
T
h
ese
ch
ar
ac
ter
is
tics
ca
n
in
clu
d
e
in
f
o
r
m
atio
n
s
u
ch
as
am
p
litu
d
e,
s
lo
p
e,
an
d
v
ar
ian
c
e.
On
ce
ea
ch
s
eg
m
en
t
is
r
ep
r
esen
ted
b
y
a
s
y
m
b
o
l,
a
T
E
SP
AR
m
atr
ix
is
co
n
s
tr
u
cted
.
T
h
is
m
atr
ix
is
a
co
m
p
ac
t
r
ep
r
esen
tatio
n
o
f
th
e
o
r
ig
in
al
s
ig
n
al,
wh
er
e
ea
ch
r
o
w
r
ep
r
esen
ts
a
s
eg
m
en
t
o
f
tim
e
an
d
ea
ch
co
lu
m
n
r
ep
r
esen
ts
a
s
y
m
b
o
l
ass
ig
n
ed
to
th
at
s
eg
m
e
n
t.
T
h
e
T
E
SP
AR
m
atr
ix
ca
n
th
e
n
b
e
u
s
ed
f
o
r
s
ig
n
al
an
aly
s
is
an
d
r
ec
o
g
n
itio
n
.
T
h
e
s
tep
s
o
f
t
h
e
T
E
SP
AR
co
d
in
g
m
o
d
el
ar
e
illu
s
tr
ated
in
th
e
f
o
llo
win
g
F
ig
u
r
e
2
.
Fig
u
r
e
2
.
T
E
SP
AR
co
d
in
g
m
o
d
el
c.
Ad
v
an
tag
es
:
t
h
e
ad
v
an
tag
es
o
f
th
is
ap
p
r
o
ac
h
ar
e
s
u
m
m
ar
ize
d
in
th
e
f
o
llo
win
g
n
o
tes:
t
em
p
o
r
al
an
d
r
ea
l
-
tim
e
p
r
o
ce
s
s
in
g
,
f
ast
e
f
f
icien
t
an
aly
s
is
an
d
r
ec
o
g
n
itio
n
,
n
o
is
e
f
ilter
in
g
,
s
im
p
licity
an
d
p
r
o
b
lem
r
ed
u
ctio
n
,
an
d
i
n
teg
r
atio
n
in
to
s
m
ar
t c
ar
d
s
(
d
ig
ital sig
n
al
p
r
o
ce
s
s
o
r
(
DSP
)
p
r
o
ce
s
s
o
r
s
)
.
d.
Dis
ad
v
an
tag
es
(
l
im
its
)
:
t
h
is
cl
ass
if
ier
s
h
o
ws
s
o
m
e
wea
k
p
o
in
ts
(
T
ab
le
1
)
s
u
ch
as:
l
ac
k
o
f
a
f
r
eq
u
e
n
cy
asp
ec
t
an
d
n
o
n
-
lin
ea
r
.
3
.
4
.
Co
m
pa
riso
n o
f
cl
a
s
s
if
iers
T
h
is
s
ec
tio
n
p
r
esen
ts
a
co
m
p
ar
is
o
n
b
etwe
en
th
e
VQ
an
d
T
E
SP
AR
cla
s
s
if
ier
s
b
as
ed
o
n
th
e
alg
o
r
ith
m
s
u
s
ed
an
d
th
e
e
v
alu
atio
n
cr
iter
ia.
a.
Alg
o
r
ith
m
s
u
s
ed
:
VQ
u
s
es
th
e
K
-
m
ea
n
s
alg
o
r
ith
m
an
d
T
E
SP
AR
u
s
es
th
e
K
-
n
ea
r
est
n
ei
g
h
b
o
r
(
KNN)
alg
o
r
ith
m
.
b.
C
o
m
p
ar
is
o
n
cr
iter
ia
:
t
h
e
cr
ite
r
ia
an
d
th
e
c
o
m
p
ar
is
o
n
r
esu
lt
b
etwe
en
V
Q
a
n
d
T
E
SP
AR
tech
n
iq
u
es
ar
e
s
h
o
w
n
in
T
ab
le
1
.
3
.
5
.
H
y
bridi
za
t
io
n
3
.
5
.
1
.
Appro
a
ches us
ed
R
ec
en
t
m
eth
o
d
s
h
a
v
e
em
e
r
g
ed
to
en
h
an
ce
th
e
r
o
b
u
s
tn
ess
o
f
r
ec
o
g
n
itio
n
s
y
s
tem
s
.
T
h
eir
c
o
m
m
o
n
ality
lies
in
th
e
u
s
e
o
f
v
ar
io
u
s
tech
n
iq
u
es
co
m
b
in
e
d
to
ar
r
i
v
e
at
a
f
in
al
d
ec
is
io
n
.
On
e
o
f
th
ese
ap
p
r
o
ac
h
es
is
th
e
u
s
e
o
f
m
u
ltip
le
class
if
ier
s
,
wh
ic
h
in
v
o
l
v
es
ap
p
ly
i
n
g
v
a
r
io
u
s
class
if
ier
s
to
th
e
s
am
e
s
p
ee
ch
s
ig
n
al,
with
th
e
d
ec
is
io
n
b
ein
g
b
ased
o
n
t
h
e
r
ec
o
m
b
in
atio
n
o
f
th
e
s
co
r
es
o
b
tain
ed
with
ea
ch
tec
h
n
iq
u
e
.
T
h
ese
ap
p
r
o
ac
h
es
em
p
h
asize
th
e
s
tan
d
ar
d
izatio
n
o
f
th
e
s
co
r
es o
b
tain
ed
f
r
o
m
e
ac
h
r
ec
o
g
n
itio
n
s
y
s
tem
,
as we
ll a
s
th
e
s
elec
tio
n
o
f
ef
f
icien
t h
y
b
r
id
izatio
n
s
tr
ateg
i
es.
3
.
5
.
2
.
So
m
e
e
x
is
t
ing
hy
bridi
z
a
t
io
ns
C
o
m
p
ar
in
g
th
e
r
esu
lts
o
f
th
es
e
d
if
f
er
en
t
h
y
b
r
id
izatio
n
s
is
co
m
p
lex
,
b
ec
au
s
e
th
e
d
atab
ase
s
,
lib
r
ar
ies,
ex
tr
ac
to
r
s
u
s
ed
an
d
ev
alu
at
io
n
cr
iter
ia
d
if
f
er
g
r
ea
tly
f
r
o
m
o
n
e
h
y
b
r
id
izatio
n
to
an
o
th
er
.
Als
o
,
th
e
ch
ar
ac
ter
is
tics
o
f
th
e
s
o
u
n
d
s
ig
n
als u
s
ed
an
d
t
h
e
f
ilter
in
g
an
d
s
eg
m
en
tatio
n
m
et
h
o
d
s
ar
e
al
s
o
s
p
ec
if
ied
.
Nev
er
th
eless
,
we
ca
n
lis
t
an
d
d
ec
lar
e
am
o
n
g
t
h
ese
wo
r
k
s
th
o
s
e
wh
ich
h
av
e
p
r
esen
ted
e
n
co
u
r
a
g
in
g
r
esu
lts
s
u
ch
as:
DT
W
-
VQ,
HM
M
-
VQ,
GM
M
-
VQ,
T
E
S
PAR
-
ANN,
an
d
AN
N
-
VQ.
T
h
is
co
llectio
n
o
f
co
m
b
in
atio
n
s
o
r
h
y
b
r
id
izatio
n
s
is
n
o
t
ex
h
au
s
tiv
e.
Oth
er
m
eth
o
d
s
ex
is
t,
b
u
t
th
ey
a
r
e
m
o
r
e
s
p
ec
if
ic
to
p
ar
ticu
lar
ca
s
es a
n
d
m
o
r
e
o
r
le
s
s
im
p
o
r
tan
t.
Evaluation Warning : The document was created with Spire.PDF for Python.
T
E
L
KOM
NI
KA
T
elec
o
m
m
u
n
C
o
m
p
u
t E
l Co
n
tr
o
l
Time
en
co
d
ed
s
ig
n
a
l p
r
o
ce
s
s
in
g
a
n
d
r
ec
o
g
n
itio
n
w
ith
ve
cto
r
q
u
a
n
tiz
a
tio
n
:
…
(
A
b
d
elma
ji
d
La
mka
d
a
m
)
485
4.
M
E
T
H
O
DO
L
O
G
Y
AND
W
O
RK
4
.
1
.
Co
rpus
a
nd
t
he
t
est
ba
s
e
T
ab
le
2
lis
ts
th
e
ac
o
u
s
tic
ch
a
r
ac
ter
is
tics
r
eq
u
ir
ed
to
c
r
ea
te
th
e
r
ef
e
r
en
ce
c
o
r
p
u
s
f
o
r
t
h
e
1
0
Ar
ab
ic
n
u
m
er
als
(
0
to
9
)
.
A
r
ec
o
r
d
in
g
o
f
th
e
c
o
r
p
u
s
was
m
ad
e
f
o
r
1
0
Mo
r
o
cc
an
s
p
ea
k
er
s
.
E
ac
h
d
ig
it
in
th
e
b
ase
{0
-
9
}
is
r
ec
o
r
d
ed
1
0
tim
es,
an
d
test
ed
1
0
tim
es
b
y
1
0
s
p
ea
k
er
s
.
All
th
e
te
s
ts
ar
e
ca
r
r
ied
o
u
t
u
s
in
g
th
e
m
el
-
f
r
eq
u
e
n
cy
ce
p
s
tr
al
co
ef
f
icien
t
s
(
MFC
C
)
ex
tr
ac
to
r
in
teg
r
ate
d
in
to
T
E
SP
AR
.
On
ce
all
th
e
s
am
p
les
h
av
e
b
ee
n
an
aly
ze
d
,
it
is
f
air
ly
s
tr
aig
h
tf
o
r
war
d
to
co
m
p
a
r
e
th
em
,
o
r
m
o
r
e
p
r
ec
is
ely
to
m
ea
s
u
r
e
th
e
s
im
ilar
ity
o
f
th
e
s
am
p
le
to
b
e
id
en
tifie
d
in
r
elat
io
n
to
th
e
r
ef
er
en
ce
s
am
p
les.
T
ab
le
2
.
Aco
u
s
tic
ch
ar
ac
ter
is
tics
o
f
s
p
ee
ch
co
r
p
u
s
P
a
r
a
me
t
e
r
V
a
l
u
e
F
o
r
mat
M
o
n
o
(
.
w
a
v
)
S
a
mp
l
i
n
g
8
kH
z
C
o
d
a
g
e
1
6
b
i
t
s
F
r
a
mes
n
u
mb
e
r
50
R
e
c
o
r
d
i
n
g
t
i
me
5
s
e
c
o
n
d
e
s
/
d
i
g
i
t
W
i
n
d
o
w
i
n
g
H
a
mm
i
n
g
4
.
2
.
So
f
t
wa
re
a
nd
t
o
o
ls
us
ed
Au
d
ac
ity
was
u
s
ed
as
a
to
o
l
f
o
r
ac
q
u
ir
in
g
an
d
an
aly
zin
g
s
o
u
n
d
f
iles
,
wh
ile
M
AT
L
AB
s
o
f
twar
e
was
ch
o
s
en
f
o
r
t
h
e
p
r
o
ce
s
s
in
g
en
v
ir
o
n
m
e
n
t,
as
it
o
f
f
er
s
n
u
m
e
r
o
u
s
s
ig
n
al
p
r
o
ce
s
s
in
g
f
u
n
cti
o
n
s
(
m
-
f
iles
)
.
T
h
is
lan
g
u
ag
e
is
also
s
im
ilar
to
th
e
n
o
tatio
n
u
s
ed
in
lin
ea
r
alg
e
b
r
a
an
d
n
u
m
er
ical
an
al
y
s
is
,
an
d
it
is
o
p
e
n
to
o
th
e
r
to
o
lb
o
x
es
f
o
r
p
er
f
o
r
m
in
g
ad
v
an
ce
d
an
d
c
o
m
p
lex
o
p
er
atio
n
s
o
n
v
ec
to
r
s
an
d
m
atr
ices
r
e
p
r
esen
tin
g
ac
o
u
s
tic
ch
ar
ac
ter
is
tics
.
4
.
3
.
H
y
bridi
za
t
io
n a
pp
ro
a
c
h
T
E
SP
AR
an
d
VQ
ar
e
two
tec
h
n
iq
u
es
u
s
ed
in
s
ig
n
al
p
r
o
ce
s
s
in
g
an
d
p
atter
n
r
ec
o
g
n
itio
n
.
C
o
m
b
in
in
g
th
em
ca
n
im
p
r
o
v
e
th
e
e
f
f
icien
cy
o
f
s
ig
n
al
co
d
i
n
g
a
n
d
r
ec
o
g
n
itio
n
,
p
a
r
ticu
lar
ly
in
s
p
ee
ch
r
ec
o
g
n
itio
n
.
In
th
is
p
r
esen
t
wo
r
k
we
ar
e
o
n
ly
in
te
r
ested
in
h
y
b
r
id
izatio
n
at
th
e
l
ev
el
o
f
class
if
ier
s
.
T
h
e
p
r
o
p
o
s
ed
h
y
b
r
i
d
izatio
n
is
V
Q
-
T
E
SP
AR
.
T
h
is
h
y
b
r
id
izatio
n
p
r
esen
ts
o
u
r
n
ew
co
n
tr
i
b
u
tio
n
to
t
h
e
class
if
icatio
n
p
h
ase
o
f
th
e
s
p
ea
k
er
r
ec
o
g
n
itio
n
s
y
s
tem
,
as it ta
k
es in
to
ac
co
u
n
t sev
er
al
asp
ec
ts
o
f
s
ig
n
al
p
r
o
ce
s
s
in
g
.
4
.
4
.
H
y
bridi
za
t
io
n pro
ce
eding
4
.
4
.
1
.
P
ro
ce
s
s
ing
a
nd
ex
t
ra
ct
io
n o
f
a
co
us
t
ic
v
ec
t
o
rs
T
h
e
f
ir
s
t
s
tag
e
co
n
s
is
ts
o
f
p
r
ep
r
o
ce
s
s
in
g
th
e
s
p
ee
ch
s
ig
n
al
an
d
ex
tr
ac
tin
g
t
h
e
r
elev
a
n
t
ac
o
u
s
tic
f
ea
tu
r
es
:
(
i)
a
cq
u
i
r
e,
f
ilter
o
u
t
u
n
wan
ted
n
o
is
e
an
d
n
o
r
m
alize
th
e
v
o
ice
s
ig
n
al
a
n
d
(
ii)
e
x
tr
ac
tio
n
o
f
ac
o
u
s
tic
v
ec
to
r
s
u
s
in
g
MFC
C
(
in
teg
r
ated
in
to
T
E
SP
AR
)
.
T
h
e
p
r
o
ce
s
s
in
g
an
d
ex
t
r
ac
tio
n
p
r
o
ce
s
s
is
s
h
o
wn
in
Fig
u
r
e
3
.
Fig
u
r
e
3
.
T
r
ea
tm
en
t
an
d
ex
tr
a
ctio
n
m
o
d
el
4
.
4
.
2
.
T
E
SPAR
c
o
di
ng
At
th
is
s
tag
e,
th
e
p
r
o
ce
s
s
ed
s
i
g
n
al
is
en
c
o
d
ed
u
s
in
g
th
e
T
E
SP
AR
tech
n
iq
u
e
;
(
i)
t
r
an
s
f
o
r
m
th
e
an
alo
g
s
ig
n
al
in
to
a
s
eq
u
en
ce
o
f
T
E
SP
AR
s
y
m
b
o
ls
(
b
ased
o
n
elem
en
tar
y
wav
ef
o
r
m
s
)
a
n
d
(
ii)
e
x
tr
ac
t
k
ey
p
ar
am
eter
s
: d
u
r
atio
n
,
am
p
litu
d
e,
an
d
f
r
eq
u
en
cy
.
4
.
4
.
3
.
Vec
t
o
r
q
ua
ntiz
a
t
io
n
T
h
e
T
E
SP
AR
-
co
d
ed
in
f
o
r
m
at
io
n
is
th
e
n
p
r
o
ce
s
s
ed
u
s
in
g
v
ec
to
r
q
u
an
tizatio
n
;
(
i)
g
r
o
u
p
T
E
SP
A
R
s
eq
u
en
ce
s
in
to
f
ea
tu
r
e
v
ec
to
r
s
,
(
ii)
u
s
e
VQ
to
co
m
p
r
ess
an
d
class
if
y
th
ese
v
ec
to
r
s
,
r
ed
u
cin
g
r
ed
u
n
d
an
c
y
,
an
d
(
iii)
a
s
s
ig
n
ea
ch
v
ec
to
r
t
o
a
co
d
eb
o
o
k
o
p
tim
ize
d
f
o
r
p
atter
n
r
ec
o
g
n
itio
n
.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
1
6
9
3
-
6
9
3
0
TEL
KOM
NI
KA
T
elec
o
m
m
u
n
C
o
m
p
u
t E
l
C
o
n
tr
o
l
,
Vo
l.
24
,
No
.
2
,
Ap
r
il
20
26
:
4
8
1
-
4
8
9
486
4
.
4
.
4
.
Reduct
io
n o
f
re
presenta
t
io
n sp
a
ce
T
h
e
ex
tr
ac
ted
ac
o
u
s
tic
v
ec
to
r
s
ar
e
g
en
er
ally
o
f
lar
g
e
d
im
en
s
io
n
ality
,
wh
ich
in
cr
ea
s
es
co
m
p
u
tatio
n
al
co
m
p
lex
ity
;
(
i)
t
h
e
ac
o
u
s
tic
v
ec
to
r
s
r
esu
ltin
g
f
r
o
m
th
e
ex
tr
ac
tio
n
o
p
er
atio
n
ar
e
v
e
r
y
lar
g
e
in
q
u
an
tity
,
s
o
to
r
ed
u
ce
th
is
q
u
an
tity
an
d
k
ee
p
o
n
ly
th
e
r
elev
a
n
t
p
ar
a
m
eter
s
a
n
d
(
ii)
t
o
o
p
tim
ize
th
e
r
ec
o
g
n
i
tio
n
o
p
e
r
atio
n
with
an
ac
ce
p
tab
le
ca
lcu
latio
n
tim
e,
we
u
s
ed
th
e
L
DA
d
is
cr
im
in
a
to
r
wh
ich
aim
s
to
r
ed
u
ce
th
e
r
ep
r
esen
tatio
n
s
p
ac
e
o
f
th
ese
ac
o
u
s
tic
p
ar
a
m
eter
s
.
4
.
4
.
5
.
Rec
o
g
nitio
n a
nd
cla
s
s
if
ica
t
io
n
T
h
e
f
in
al
s
tag
e
co
r
r
esp
o
n
d
s
to
th
e
r
ec
o
g
n
itio
n
an
d
class
if
icatio
n
p
r
o
ce
s
s
;
(
i)
a
p
p
l
y
m
ac
h
i
n
e
lear
n
in
g
tech
n
iq
u
es
s
u
ch
as
to
id
en
tif
y
p
atter
n
s
an
d
(
ii)
c
o
m
p
ar
e
t
h
e
VQ
s
to
a
d
atab
ase
o
f
k
n
o
w
n
p
atter
n
s
.
Fig
u
r
e
4
s
h
o
ws th
e
h
y
b
r
i
d
izatio
n
p
h
ases
o
f
th
e
two
m
eth
o
d
s
V
Q
an
d
T
E
SP
AR
.
Fig
u
r
e
4
.
V
Q
an
d
T
E
SP
AR
h
y
b
r
id
izatio
n
m
o
d
el
4
.
5
.
E
x
pla
na
t
io
n
I
n
th
e
T
E
SP
AR
tech
n
iq
u
e,
r
e
f
er
en
ce
a
n
d
test
m
o
d
els
a
r
e
r
ep
r
esen
ted
as
m
atr
ices,
alig
n
e
d
in
tim
e,
g
en
er
atin
g
a
s
co
r
e
u
s
ed
f
o
r
r
ec
o
g
n
itio
n
.
W
h
er
ea
s
in
th
e
V
Q
m
eth
o
d
,
r
ef
er
en
ce
m
o
d
els
ar
e
r
ep
r
esen
ted
b
y
co
d
e
s
eq
u
en
ce
s
,
wh
ile
test
m
o
d
els
ar
e
illu
s
tr
ated
b
y
s
p
ec
tr
al
p
ar
am
eter
s
eq
u
e
n
ce
s
.
An
a
v
er
ag
e
q
u
an
tizatio
n
d
is
to
r
tio
n
is
ev
alu
ated
o
n
th
e
r
ef
er
en
ce
m
o
d
els
f
o
r
ea
c
h
test
in
s
tr
u
ctio
n
,
an
d
th
is
ev
alu
atio
n
is
th
en
u
s
ed
with
th
e
d
ec
is
io
n
th
r
esh
o
ld
in
th
e
v
er
if
icatio
n
p
r
o
ce
s
s
.
B
y
u
s
i
n
g
th
e
V
Q
tech
n
iq
u
e
f
o
r
th
is
r
ep
r
esen
tatio
n
,
th
e
am
o
u
n
t
o
f
ca
lcu
latio
n
s
a
n
d
s
to
r
ag
e
is
co
n
s
id
er
a
b
ly
r
ed
u
ce
d
.
Fin
ally
,
t
h
e
DT
W
m
eth
o
d
is
ca
lled
to
o
b
tain
a
s
co
r
e
(
m
in
im
u
m
d
is
tan
ce
)
b
et
wee
n
th
e
r
ef
er
e
n
ce
an
d
test
VQ
co
d
es,
wh
ich
will b
e
u
s
ed
f
o
r
th
e
f
i
n
al
d
ec
is
io
n
.
5.
RE
SU
L
T
S
AND
D
I
SCU
SS
I
O
N
C
o
m
p
ar
in
g
a
s
in
g
le
d
ig
it
(
p
r
o
n
o
u
n
ce
d
in
Ar
ab
ic)
with
s
ev
er
al
d
ig
its
in
th
e
n
u
m
b
er
b
ase
(
0
to
9
)
led
to
th
e
f
o
llo
win
g
co
m
p
ar
ativ
e
r
esu
lt
(
T
ab
le
3
an
d
Fig
u
r
e
5
)
.
T
h
e
av
er
a
g
e
r
esu
lt
(
T
ab
le
4
an
d
Fig
u
r
e
6
)
f
o
r
r
ec
o
g
n
izin
g
th
e
1
0
Ar
ab
ic
n
u
m
er
als is
as f
o
llo
ws:
−
T
E
SP
AR
r
ec
o
g
n
itio
n
(
6
7
%)
w
as sl
ig
h
tly
b
etter
th
an
VQ
(
6
2
%)
−
T
h
e
n
ew
h
y
b
r
id
izatio
n
s
h
o
we
d
m
ea
s
u
r
ab
le
in
c
r
ea
s
es in
th
e
r
ec
o
g
n
itio
n
r
ate
(
7
2
.
5
%)
−
R
ec
o
g
n
itio
n
b
y
th
e
n
ew
h
y
b
r
id
izatio
n
(
VQ
-
T
E
SP
AR
)
is
b
etter
th
an
t
h
e
VQ
a
n
d
T
E
SP
AR
tech
n
iq
u
es
u
s
ed
in
d
iv
id
u
ally
f
o
r
ea
ch
d
ig
i
t in
th
e
co
r
p
u
s
(
0
-
9
)
(
T
ab
le
3
a
n
d
Fig
u
r
e
5
)
To
o
b
jectiv
ely
co
m
p
ar
e
th
e
r
e
s
u
lts
o
f
th
is
wo
r
k
with
o
th
er
h
y
b
r
id
izatio
n
s
s
u
ch
as
T
E
SP
A
R
-
SVM
o
r
T
E
SP
AR
-
HM
M,
wh
ich
ar
e
s
ti
ll in
d
ev
elo
p
m
en
t,
th
e
s
am
e
c
o
r
p
u
s
m
u
s
t b
e
u
s
ed
u
n
d
er
th
e
s
a
m
e
co
n
d
itio
n
s
.
T
ab
le
3
.
Dig
it
r
ec
o
g
n
itio
n
r
ate
u
s
in
g
VQ,
T
E
SP
AR
m
eth
o
d
s
an
d
th
e
h
y
b
r
i
d
izatio
n
p
e
r
f
o
r
m
ed
D
i
g
i
t
V
Q
(
%)
TESP
A
R
(
%)
H
y
b
r
i
d
i
z
a
t
i
o
n
(
%)
0
(
S
i
f
f
r
)
80
80
90
1
(
W
a
h
e
d
)
70
75
80
2
(
I
t
h
n
a
n
)
60
70
75
3
(
T
h
a
l
a
t
h
a
)
40
50
50
4
(
A
r
b
a
a
)
60
70
70
5
(
K
h
a
msa)
50
60
70
6
(
S
i
t
t
a
)
80
80
90
7
(
S
a
b
a
a
)
50
60
60
8
(
T
h
a
m
a
n
i
a
)
70
75
80
9
(
T
i
saa
)
60
50
60
Evaluation Warning : The document was created with Spire.PDF for Python.
T
E
L
KOM
NI
KA
T
elec
o
m
m
u
n
C
o
m
p
u
t E
l Co
n
tr
o
l
Time
en
co
d
ed
s
ig
n
a
l p
r
o
ce
s
s
in
g
a
n
d
r
ec
o
g
n
itio
n
w
ith
ve
cto
r
q
u
a
n
tiz
a
tio
n
:
…
(
A
b
d
elma
ji
d
La
mka
d
a
m
)
487
Fig
u
r
e
5
.
I
ll
u
s
tr
atio
n
(
co
m
p
ar
i
s
o
n
)
o
f
d
ig
it r
ec
o
g
n
itio
n
r
ate
r
esu
lts
T
ab
le
4
.
Av
e
r
ag
e
r
ec
o
g
n
itio
n
r
ate
u
s
in
g
V
Q
,
T
E
SP
AR
m
eth
o
d
s
an
d
th
e
h
y
b
r
id
izatio
n
p
er
f
o
r
m
ed
Te
c
h
n
i
q
u
e
V
Q
(
%)
TESP
A
R
(
%)
H
y
b
r
i
d
i
z
a
t
i
o
n
(
%)
A
v
e
r
a
g
e
62
67
7
2
.
5
Fig
u
r
e
6
.
I
ll
u
s
tr
atio
n
(
co
m
p
ar
i
s
o
n
)
o
f
a
v
er
ag
e
r
ec
o
g
n
itio
n
r
at
e
r
esu
lts
6.
CO
NCLU
SI
O
N
At
th
e
b
eg
in
n
in
g
o
f
th
is
wo
r
k
,
we
d
em
o
n
s
tr
ated
th
e
p
r
o
ce
s
s
in
g
p
r
o
ce
s
s
es
o
f
th
e
two
cla
s
s
if
icatio
n
m
eth
o
d
s
,
V
Q
an
d
T
E
SP
AR
,
as
well
as
th
eir
al
g
o
r
ith
m
s
.
W
e
co
m
p
ar
ed
th
ese
two
class
if
ier
s
b
ased
o
n
th
e
m
o
s
t
well
-
k
n
o
wn
an
d
co
m
m
o
n
ly
u
s
ed
asp
ec
ts
o
f
s
p
ee
ch
p
r
o
ce
s
s
in
g
.
T
h
en
we
cited
th
e
v
ar
io
u
s
h
y
b
r
id
ized
s
y
s
tem
s
alr
ea
d
y
d
ev
elo
p
ed
o
v
er
th
e
p
a
s
t
d
ec
ad
e.
I
n
d
ee
d
,
th
ese
s
tu
d
ies,
d
ed
icate
d
to
h
y
b
r
id
izin
g
ce
r
tain
m
eth
o
d
s
,
h
av
e
p
r
o
d
u
ce
d
im
p
r
o
v
ed
an
d
ef
f
ici
en
t
r
esu
lts
.
T
h
is
,
o
f
co
u
r
s
e,
le
d
to
th
e
i
d
ea
o
f
p
r
o
p
o
s
in
g
a
h
y
b
r
id
m
eth
o
d
b
ased
o
n
two
e
x
is
tin
g
tech
n
iq
u
es,
to
in
cr
ea
s
e
th
e
lev
el
o
f
c
o
m
p
ar
i
s
o
n
an
d
r
aise
th
e
q
u
ality
o
f
r
e
co
g
n
itio
n
.
W
ith
th
e
h
y
b
r
id
izatio
n
ac
h
ie
v
ed
at
th
is
lev
el,
we
h
av
e
ac
h
iev
e
d
e
n
c
o
u
r
ag
in
g
r
esu
lts
,
p
av
in
g
th
e
way
f
o
r
in
n
o
v
ativ
e
co
m
b
in
atio
n
s
a
n
d
h
y
b
r
id
izati
o
n
s
lik
e
with
DT
W
,
SVM,
an
d
HM
M
.
A
s
tu
d
y
o
n
t
h
e
p
o
s
s
ib
ilit
y
o
f
ap
p
ly
in
g
th
is
wo
r
k
to
s
m
ar
t
ca
r
d
s
co
u
ld
b
e
p
lan
n
ed
,
in
o
r
d
e
r
to
cr
ea
te
a
r
ea
l
m
in
i
v
o
ice
au
th
en
ticatio
n
s
y
s
tem
b
ased
o
n
s
m
ar
t c
ar
d
s
.
ACK
NO
WL
E
DG
M
E
N
T
S
We
wo
u
ld
lik
e
to
th
a
n
k
t
h
e
L
E
MSA
L
ab
o
r
ato
r
y
o
f
th
e
Fac
u
lty
o
f
Scien
ce
s
,
Fez,
M
o
r
o
cc
o
,
f
o
r
th
e
r
eso
u
r
ce
s
m
ad
e
a
v
ailab
le
to
u
s
f
o
r
th
is
wo
r
k
.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
1
6
9
3
-
6
9
3
0
TEL
KOM
NI
KA
T
elec
o
m
m
u
n
C
o
m
p
u
t E
l
C
o
n
tr
o
l
,
Vo
l.
24
,
No
.
2
,
Ap
r
il
20
26
:
4
8
1
-
4
8
9
488
F
UNDING
I
NF
O
R
M
A
T
I
O
N
Au
th
o
r
s
s
tate
n
o
f
u
n
d
in
g
in
v
o
lv
ed
.
AUTHO
R
CO
NT
RI
B
UT
I
O
NS ST
A
T
E
M
E
N
T
T
h
is
jo
u
r
n
al
u
s
es
th
e
C
o
n
tr
ib
u
to
r
R
o
les
T
ax
o
n
o
m
y
(
C
R
ed
iT
)
to
r
ec
o
g
n
ize
in
d
iv
id
u
al
au
th
o
r
co
n
tr
ib
u
tio
n
s
,
r
ed
u
ce
au
th
o
r
s
h
ip
d
is
p
u
tes,
an
d
f
ac
ilit
ate
co
llab
o
r
atio
n
.
Na
m
e
o
f
Aut
ho
r
C
M
So
Va
Fo
I
R
D
O
E
Vi
Su
P
Fu
Ab
d
elm
ajid
L
am
k
ad
am
✓
✓
✓
✓
✓
✓
✓
✓
✓
✓
✓
Mo
h
am
m
ed
Kar
i
m
✓
✓
✓
✓
✓
C
:
C
o
n
c
e
p
t
u
a
l
i
z
a
t
i
o
n
M
:
M
e
t
h
o
d
o
l
o
g
y
So
:
So
f
t
w
a
r
e
Va
:
Va
l
i
d
a
t
i
o
n
Fo
:
Fo
r
mal
a
n
a
l
y
s
i
s
I
:
I
n
v
e
s
t
i
g
a
t
i
o
n
R
:
R
e
so
u
r
c
e
s
D
:
D
a
t
a
C
u
r
a
t
i
o
n
O
:
W
r
i
t
i
n
g
-
O
r
i
g
i
n
a
l
D
r
a
f
t
E
:
W
r
i
t
i
n
g
-
R
e
v
i
e
w
&
E
d
i
t
i
n
g
Vi
:
Vi
su
a
l
i
z
a
t
i
o
n
Su
:
Su
p
e
r
v
i
s
i
o
n
P
:
P
r
o
j
e
c
t
a
d
mi
n
i
st
r
a
t
i
o
n
Fu
:
Fu
n
d
i
n
g
a
c
q
u
i
si
t
i
o
n
CO
NF
L
I
C
T
O
F
I
N
T
E
R
E
S
T
ST
A
T
E
M
E
NT
Au
th
o
r
s
s
tate
n
o
co
n
f
lict o
f
in
t
er
est.
DATA AV
AI
L
AB
I
L
I
T
Y
Data
av
ailab
ilit
y
is
n
o
t
a
p
p
li
ca
b
le
to
th
is
p
ap
er
as
n
o
n
e
w
d
ata
wer
e
c
r
ea
ted
o
r
an
al
y
ze
d
in
t
h
is
s
tu
d
y
.
RE
F
E
R
E
NC
E
S
[
1
]
J.
R
.
Z
u
ma
l
t
,
“
V
o
i
c
e
r
e
c
o
g
n
i
t
i
o
n
t
e
c
h
n
o
l
o
g
y
:
H
a
s
i
t
c
o
me
o
f
a
g
e
?
,
”
I
n
f
o
rm
a
t
i
o
n
T
e
c
h
n
o
l
o
g
y
a
n
d
L
i
b
r
a
r
i
e
s
,
v
o
l
.
2
4
,
n
o
.
4
,
p
p
.
1
8
0
–
1
8
5
,
D
e
c
.
2
0
0
5
,
d
o
i
:
1
0
.
6
0
1
7
/
i
t
a
l
.
v
2
4
i
4
.
3
3
8
2
.
[
2
]
L.
Li
,
Y
.
L
i
n
,
Z.
Z
h
a
n
g
,
a
n
d
D
.
W
a
n
g
,
“
I
mp
r
o
v
e
d
d
e
e
p
s
p
e
a
k
e
r
f
e
a
t
u
r
e
l
e
a
r
n
i
n
g
f
o
r
t
e
x
t
-
d
e
p
e
n
d
e
n
t
sp
e
a
k
e
r
r
e
c
o
g
n
i
t
i
o
n
,
”
i
n
2
0
1
5
Asi
a
-
Pa
c
i
f
i
c
S
i
g
n
a
l
a
n
d
I
n
f
o
rm
a
t
i
o
n
Pr
o
c
e
ss
i
n
g
Ass
o
c
i
a
t
i
o
n
An
n
u
a
l
S
u
m
m
i
t
a
n
d
C
o
n
f
e
re
n
c
e
,
AP
S
I
PA
A
S
C
2
0
1
5
,
I
EEE,
D
e
c
.
2
0
1
6
,
p
p
.
4
2
6
–
4
2
9
,
d
o
i
:
1
0
.
1
1
0
9
/
A
P
S
I
P
A
.
2
0
1
5
.
7
4
1
5
3
0
6
.
[
3
]
A
.
K
o
u
r
d
a
n
d
K
.
K
o
u
r
d
,
“
A
r
a
b
i
c
I
so
l
a
t
e
d
W
o
r
d
S
p
e
a
k
e
r
D
e
p
e
n
d
e
n
t
R
e
c
o
g
n
i
t
i
o
n
S
y
st
e
m,”
Bri
t
i
s
h
J
o
u
r
n
a
l
o
f
Ma
t
h
e
m
a
t
i
c
s
&
C
o
m
p
u
t
e
r
S
c
i
e
n
c
e
,
v
o
l
.
1
4
,
n
o
.
1
,
p
p
.
1
–
1
5
,
J
a
n
.
2
0
1
6
,
d
o
i
:
1
0
.
9
7
3
4
/
b
j
m
c
s/
2
0
1
6
/
2
3
0
3
4
.
[
4
]
F
.
A
.
A
.
La
l
e
y
e
,
E.
C
.
Ez
i
n
,
a
n
d
C
.
M
o
t
a
m
e
d
,
“
A
u
t
o
mat
i
c
f
o
n
g
b
e
p
h
o
n
e
me
r
e
c
o
g
n
i
t
i
o
n
f
r
o
m
s
p
o
k
e
n
sp
e
e
c
h
s
i
g
n
a
l
,
”
i
n
I
C
I
N
C
O
2
0
1
6
-
Pro
c
e
e
d
i
n
g
s
o
f
t
h
e
1
3
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
f
o
rm
a
t
i
c
s
i
n
C
o
n
t
r
o
l
,
Au
t
o
m
a
t
i
o
n
a
n
d
R
o
b
o
t
i
c
s
,
S
C
I
TEPRE
S
S
-
S
c
i
e
n
c
e
a
n
d
T
e
c
h
n
o
l
o
g
y
P
u
b
l
i
c
a
t
i
o
n
s
,
2
0
1
6
,
p
p
.
1
0
2
–
1
0
9
,
d
o
i
:
1
0
.
5
2
2
0
/
0
0
0
6
0
0
4
1
0
1
0
2
0
1
0
9
.
[
5
]
R
.
C
h
a
r
a
n
,
A
.
M
a
n
i
sh
a
,
R
.
K
a
r
t
h
i
k
,
a
n
d
M
.
R
.
K
u
m
a
r
,
“
A
t
e
x
t
-
i
n
d
e
p
e
n
d
e
n
t
s
p
e
a
k
e
r
v
e
r
i
f
i
c
a
t
i
o
n
m
o
d
e
l
:
A
c
o
mp
a
r
a
t
i
v
e
a
n
a
l
y
s
i
s,
”
i
n
P
ro
c
e
e
d
i
n
g
s
o
f
2
0
1
7
I
n
t
e
r
n
a
t
i
o
n
a
l
C
o
n
f
e
re
n
c
e
o
n
I
n
t
e
l
l
i
g
e
n
t
C
o
m
p
u
t
i
n
g
a
n
d
C
o
n
t
r
o
l
,
I
2
C
2
2
0
1
7
,
I
EEE,
Ju
n
.
2
0
1
8
,
p
p
.
1
–
6
,
d
o
i
:
1
0
.
1
1
0
9
/
I
2
C
2
.
2
0
1
7
.
8
3
2
1
7
9
4
.
[
6
]
A
.
M
.
A
h
m
e
d
a
n
d
A
.
K
.
H
a
s
sa
n
,
“
S
p
e
a
k
e
r
R
e
c
o
g
n
i
t
i
o
n
S
y
s
t
e
m
s
i
n
t
h
e
L
a
st
D
e
c
a
d
e
–
A
S
u
r
v
e
y
,
”
En
g
i
n
e
e
ri
n
g
a
n
d
T
e
c
h
n
o
l
o
g
y
J
o
u
rn
a
l
,
v
o
l
.
3
9
,
n
o
.
1
B
,
p
p
.
3
0
–
4
0
,
M
a
r
.
2
0
2
1
,
d
o
i
:
1
0
.
3
0
6
8
4
/
e
t
j
.
v
3
9
i
1
b
.
1
5
8
9
.
[
7
]
A
.
B
r
e
w
a
n
d
P
.
C
u
n
n
i
n
g
h
a
m
,
“
V
e
c
t
o
r
q
u
a
n
t
i
sa
t
i
o
n
ma
p
p
i
n
g
s
f
o
r
s
p
e
a
k
e
r
v
e
r
i
f
i
c
a
t
i
o
n
,
”
i
n
Pr
o
c
e
e
d
i
n
g
s
-
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
P
a
t
t
e
r
n
R
e
c
o
g
n
i
t
i
o
n
,
I
EEE,
A
u
g
.
2
0
1
0
,
p
p
.
5
6
0
–
5
6
4
,
d
o
i
:
1
0
.
1
1
0
9
/
I
C
P
R
.
2
0
1
0
.
1
4
2
.
[
8
]
M
.
K
.
G
i
l
l
,
R
.
K
a
u
r
,
a
n
d
J
.
K
a
u
r
,
“
V
e
c
t
o
r
Q
u
a
n
t
i
z
a
t
i
o
n
b
a
se
d
S
p
e
a
k
e
r
I
d
e
n
t
i
f
i
c
a
t
i
o
n
,
”
I
n
t
e
r
n
a
t
i
o
n
a
l
J
o
u
r
n
a
l
o
f
C
o
m
p
u
t
e
r
Ap
p
l
i
c
a
t
i
o
n
s
,
v
o
l
.
4
,
n
o
.
2
,
p
p
.
1
–
4
,
J
u
l
.
2
0
1
0
,
d
o
i
:
1
0
.
5
1
2
0
/
8
0
6
-
1
1
4
6
.
[
9
]
D
.
H
a
y
e
t
a
n
d
L
.
M
.
Ta
y
e
b
,
“
U
si
n
g
v
e
c
t
o
r
q
u
a
n
t
i
z
a
t
i
o
n
f
o
r
u
n
i
v
e
r
sa
l
b
a
c
k
g
r
o
u
n
d
m
o
d
e
l
i
n
a
u
t
o
ma
t
i
c
s
p
e
a
k
e
r
v
e
r
i
f
i
c
a
t
i
o
n
,
”
i
n
C
EU
R
Wo
r
k
s
h
o
p
Pro
c
e
e
d
i
n
g
s
,
2
0
1
2
,
p
p
.
1
1
2
–
1
2
0
.
[
1
0
]
B
.
Q
.
A
l
-
A
b
u
d
i
a
n
d
M
.
S
.
M
a
h
d
i
,
“
S
p
e
a
k
e
r
A
u
t
h
e
n
t
i
c
a
t
i
o
n
U
si
n
g
V
e
c
t
o
r
Q
u
a
n
t
i
z
a
t
i
o
n
,
”
Ba
g
h
d
a
d
S
c
i
e
n
c
e
J
o
u
r
n
a
l
,
v
o
l
.
6
,
n
o
.
4
,
p
p
.
8
0
4
–
8
1
0
,
J
a
n
.
2
0
2
5
,
d
o
i
:
1
0
.
2
1
1
2
3
/
b
sj
.
2
0
0
9
.
1
1
8
8
8
.
[
1
1
]
E.
L
u
p
u
,
Z.
F
e
h
e
r
,
a
n
d
P
.
G
.
P
o
p
,
“
O
n
t
h
e
s
p
e
a
k
e
r
v
e
r
i
f
i
c
a
t
i
o
n
u
s
i
n
g
t
h
e
TESP
A
R
c
o
d
i
n
g
me
t
h
o
d
,
”
i
n
S
C
S
2
0
0
3
.
I
n
t
e
rn
a
t
i
o
n
a
l
S
y
m
p
o
s
i
u
m
o
n
S
i
g
n
a
l
s,
C
i
rc
u
i
t
s
a
n
d
S
y
s
t
e
m
s
.
Pr
o
c
e
e
d
i
n
g
s
(
C
a
t
.
N
o
.
0
3
E
X7
2
0
)
,
I
EEE,
p
p
.
1
7
3
–
176
,
d
o
i
:
1
0
.
1
1
0
9
/
S
C
S
.
2
0
0
3
.
1
2
2
6
9
7
6
.
[
1
2
]
K
.
S
.
P
r
a
s
a
d
,
K
.
A
.
S
h
e
e
l
a
,
a
n
d
M
.
S
r
i
d
e
v
i
,
“
O
p
t
i
m
i
z
a
t
i
o
n
o
f
TESP
A
R
f
e
a
t
u
r
e
s
u
s
i
n
g
r
o
b
u
s
t
F
-
r
a
t
i
o
f
o
r
s
p
e
a
k
e
r
r
e
c
o
g
n
i
t
i
o
n
,
”
i
n
Pro
c
e
e
d
i
n
g
s
o
f
I
C
S
C
N
2
0
0
7
:
I
n
t
e
r
n
a
t
i
o
n
a
l
C
o
n
f
e
re
n
c
e
o
n
S
i
g
n
a
l
Pr
o
c
e
ssi
n
g
C
o
m
m
u
n
i
c
a
t
i
o
n
s
a
n
d
N
e
t
w
o
rk
i
n
g
,
I
EEE,
F
e
b
.
2
0
0
7
,
p
p
.
2
0
–
25
,
d
o
i
:
1
0
.
1
1
0
9
/
I
C
S
C
N
.
2
0
0
7
.
3
5
0
6
7
3
.
[
1
3
]
P
.
O
.
A
d
a
r
a
n
d
M
.
O
.
O
d
h
i
a
m
b
o
,
“
TE
S
P
A
R
c
o
d
e
d
s
p
e
e
c
h
q
u
a
l
i
t
y
e
v
a
l
u
a
t
i
o
n
(
TC
S
Q
E)
,
”
i
n
AFRI
C
O
N
2
0
0
9
,
I
EE
E,
S
e
p
.
2
0
0
9
,
p
p
.
1
–
6
,
d
o
i
:
1
0
.
1
1
0
9
/
A
F
R
C
O
N
.
2
0
0
9
.
5
3
0
8
1
9
8
.
[
1
4
]
G
.
Y
o
u
a
n
d
Y
.
Li
,
“
En
v
i
r
o
n
me
n
t
a
l
s
o
u
n
d
s
r
e
c
o
g
n
i
t
i
o
n
u
si
n
g
TESP
A
R
,
”
i
n
2
0
1
2
5
t
h
I
n
t
e
r
n
a
t
i
o
n
a
l
C
o
n
g
r
e
ss
o
n
I
m
a
g
e
a
n
d
S
i
g
n
a
l
Pro
c
e
ssi
n
g
,
C
I
S
P
2
0
1
2
,
I
EEE,
O
c
t
.
2
0
1
2
,
p
p
.
1
7
9
6
–
1
8
0
0
,
d
o
i
:
1
0
.
1
1
0
9
/
C
I
S
P
.
2
0
1
2
.
6
4
6
9
7
8
1
.
[
1
5
]
G
.
P
.
P
o
p
,
E.
L
u
p
u
,
a
n
d
E.
O
n
a
c
a
,
“
P
a
t
h
o
l
o
g
i
c
a
l
V
o
i
c
e
A
ss
e
ssm
e
n
t
u
si
n
g
TESP
A
R
A
n
a
l
y
si
s,
”
2
0
1
4
,
p
p
.
2
3
7
–
240
,
d
o
i
:
1
0
.
1
0
0
7
/
9
7
8
-
3
-
3
1
9
-
0
7
6
5
3
-
9
_
4
8
.
[
1
6
]
G
.
P
.
P
o
p
,
“
D
i
scri
m
i
n
a
t
e
a
n
i
mal
s
o
u
n
d
s
u
s
i
n
g
T
ESP
A
R
a
n
a
l
y
si
s,
”
i
n
I
FM
B
E
Pro
c
e
e
d
i
n
g
s
,
2
0
1
7
,
p
p
.
1
8
5
–
1
8
8
,
d
o
i
:
1
0
.
1
0
0
7
/
9
7
8
-
Evaluation Warning : The document was created with Spire.PDF for Python.
T
E
L
KOM
NI
KA
T
elec
o
m
m
u
n
C
o
m
p
u
t E
l Co
n
tr
o
l
Time
en
co
d
ed
s
ig
n
a
l p
r
o
ce
s
s
in
g
a
n
d
r
ec
o
g
n
itio
n
w
ith
ve
cto
r
q
u
a
n
tiz
a
tio
n
:
…
(
A
b
d
elma
ji
d
La
mka
d
a
m
)
489
3
-
3
1
9
-
5
2
8
7
5
-
5
_
4
1
.
[
1
7
]
P
.
K
.
K
u
r
z
e
k
a
r
,
R
.
R
.
D
e
s
h
m
u
k
h
,
V
.
B
.
W
a
g
h
mare
,
a
n
d
P
.
P
.
S
h
r
i
s
h
r
i
mal
,
“
A
C
o
m
p
a
r
a
t
i
v
e
S
t
u
d
y
o
f
F
e
a
t
u
r
e
E
x
t
r
a
c
t
i
o
n
Te
c
h
n
i
q
u
e
s
f
o
r
S
p
e
e
c
h
R
e
c
o
g
n
i
t
i
o
n
S
y
st
e
m
,
”
I
n
t
e
r
n
a
t
i
o
n
a
l
J
o
u
r
n
a
l
o
f
I
n
n
o
v
a
t
i
v
e
Re
se
a
r
c
h
i
n
S
c
i
e
n
c
e
,
En
g
i
n
e
e
ri
n
g
a
n
d
T
e
c
h
n
o
l
o
g
y
,
v
o
l
.
0
3
,
n
o
.
1
2
,
p
p
.
1
8
0
0
6
–
1
8
0
1
6
,
D
e
c
.
2
0
1
4
,
d
o
i
:
1
0
.
1
5
6
8
0
/
i
j
i
r
set
.
2
0
1
4
.
0
3
1
2
0
3
4
.
[
1
8
]
J
.
K
.
S
a
n
d
h
u
,
“
A
C
o
m
p
a
r
a
t
i
v
e
S
t
u
d
y
o
f
F
e
a
t
u
r
e
E
x
t
r
a
c
t
i
o
n
a
n
d
C
l
a
ss
i
f
i
c
a
t
i
o
n
T
e
c
h
n
i
q
u
e
s
f
o
r
P
u
n
j
a
b
i
S
p
e
e
c
h
R
e
c
o
g
n
i
t
i
o
n
,
”
J
o
u
rn
a
l
o
f
I
n
f
o
rm
a
t
i
o
n
S
y
s
t
e
m
s
En
g
i
n
e
e
ri
n
g
a
n
d
M
a
n
a
g
e
m
e
n
t
,
v
o
l
.
1
0
,
n
o
.
4
6
s,
p
p
.
1
–
1
7
,
M
a
y
2
0
2
5
,
d
o
i
:
1
0
.
5
2
7
8
3
/
j
i
s
e
m.
v
1
0
i
4
6
s
.
8
7
6
9
.
[
1
9
]
L
.
J
i
n
g
j
i
a
o
,
S
.
J
i
e
,
a
n
d
L
.
Y
a
n
q
u
a
n
,
“
A
h
y
b
r
i
d
m
o
d
e
l
o
f
h
i
d
d
e
n
M
a
r
k
o
v
mo
d
e
l
s
a
n
d
a
se
l
f
-
o
r
g
a
n
i
z
i
n
g
n
e
u
r
a
l
n
e
t
w
o
r
k
m
o
d
e
l
i
n
sp
e
e
c
h
r
e
c
o
g
n
i
t
i
o
n
,
”
i
n
W
C
C
2
0
0
0
-
I
C
S
P
2
0
0
0
.
2
0
0
0
5
t
h
I
n
t
e
r
n
a
t
i
o
n
a
l
C
o
n
f
e
re
n
c
e
o
n
S
i
g
n
a
l
Pr
o
c
e
ssi
n
g
Pr
o
c
e
e
d
i
n
g
s
.
1
6
t
h
Wo
r
l
d
C
o
m
p
u
t
e
r
C
o
n
g
ress
2
0
0
0
,
I
EEE,
2
0
0
2
,
p
p
.
7
4
2
–
7
4
6
,
d
o
i
:
1
0
.
1
1
0
9
/
i
c
o
sp
.
2
0
0
0
.
8
9
1
6
1
8
.
[
2
0
]
T.
G
.
C
l
a
r
k
so
n
,
C
.
C
.
C
h
r
i
s
t
o
d
o
u
l
o
u
,
Y
.
G
u
a
n
,
D
.
G
o
r
se,
D
.
A
.
R
o
ma
n
o
-
C
r
i
t
c
h
l
e
y
,
a
n
d
J.
G
.
Ta
y
l
o
r
,
“
S
p
e
a
k
e
r
i
d
e
n
t
i
f
i
c
a
t
i
o
n
f
o
r
sec
u
r
i
t
y
sy
s
t
e
ms
u
si
n
g
r
e
i
n
f
o
r
c
e
m
e
n
t
-
t
r
a
i
n
e
d
p
R
A
M
n
e
u
r
a
l
n
e
t
w
o
r
k
a
r
c
h
i
t
e
c
t
u
r
e
s,
”
I
EE
E
T
r
a
n
sa
c
t
i
o
n
s
o
n
S
y
st
e
m
s
,
Ma
n
a
n
d
C
y
b
e
r
n
e
t
i
c
s P
a
rt
C
:
Ap
p
l
i
c
a
t
i
o
n
s
a
n
d
Re
v
i
e
w
s
,
v
o
l
.
3
1
,
n
o
.
1
,
p
p
.
6
5
–
7
6
,
2
0
0
1
,
d
o
i
:
1
0
.
1
1
0
9
/
5
3
2
6
.
9
2
3
2
6
9
.
[
2
1
]
M
.
Z.
I
l
y
a
s
,
S
.
A
.
S
a
m
a
d
,
A
.
H
u
ssa
i
n
,
a
n
d
K
.
A
.
I
sh
a
k
,
“
S
p
e
a
k
e
r
v
e
r
i
f
i
c
a
t
i
o
n
u
s
i
n
g
v
e
c
t
o
r
q
u
a
n
t
i
z
a
t
i
o
n
a
n
d
H
i
d
d
e
n
M
a
r
k
o
v
M
o
d
e
l
,
”
i
n
2
0
0
7
5
t
h
S
t
u
d
e
n
t
C
o
n
f
e
r
e
n
c
e
o
n
Re
s
e
a
rc
h
a
n
d
D
e
v
e
l
o
p
m
e
n
t
,
S
C
O
RED
,
I
EEE,
2
0
0
7
,
p
p
.
1
–
5
,
d
o
i
:
1
0
.
1
1
0
9
/
S
C
O
R
ED
.
2
0
0
7
.
4
4
5
1
4
1
9
.
[
2
2
]
M
.
K
e
s
h
a
v
a
r
z
B
a
h
a
g
h
i
g
h
a
t
,
F
.
S
a
h
b
a
,
a
n
d
E.
Te
h
r
a
n
i
,
“
T
e
x
t
-
d
e
p
e
n
d
e
n
t
S
p
e
a
k
e
r
R
e
c
o
g
n
i
t
i
o
n
b
y
C
o
m
b
i
n
a
t
i
o
n
o
f
LB
G
V
Q
a
n
d
D
TW
f
o
r
P
e
r
s
i
a
n
La
n
g
u
a
g
e
,
”
I
n
t
e
r
n
a
t
i
o
n
a
l
J
o
u
rn
a
l
o
f
C
o
m
p
u
t
e
r
A
p
p
l
i
c
a
t
i
o
n
s
,
v
o
l
.
5
1
,
n
o
.
1
6
,
p
p
.
2
3
–
2
7
,
A
u
g
.
2
0
1
2
,
d
o
i
:
1
0
.
5
1
2
0
/
8
1
2
6
-
1
7
1
1
.
[
2
3
]
R
.
B
h
a
r
t
i
a
n
d
P
.
B
a
n
sa
l
,
“
R
e
a
l
T
i
me
S
p
e
a
k
e
r
R
e
c
o
g
n
i
t
i
o
n
S
y
s
t
e
m
u
si
n
g
M
F
C
C
a
n
d
V
e
c
t
o
r
Q
u
a
n
t
i
z
a
t
i
o
n
Te
c
h
n
i
q
u
e
,
”
I
n
t
e
r
n
a
t
i
o
n
a
l
J
o
u
r
n
a
l
o
f
C
o
m
p
u
t
e
r
Ap
p
l
i
c
a
t
i
o
n
s
,
v
o
l
.
1
1
7
,
n
o
.
1
,
p
p
.
2
5
–
3
1
,
M
a
y
2
0
1
5
,
d
o
i
:
1
0
.
5
1
2
0
/
2
0
5
2
0
-
2
3
6
1
.
[
2
4
]
A
.
L
a
mk
a
d
a
m
a
n
d
M
.
K
a
r
i
m
,
“
E
n
h
a
n
c
e
d
r
e
c
o
g
n
i
t
i
o
n
o
f
A
r
a
b
i
c
n
u
m
e
r
a
l
s
b
a
s
e
d
o
n
t
h
e
c
o
mb
i
n
a
t
i
o
n
o
f
TES
P
A
R
a
n
d
M
F
C
C
e
n
c
o
d
e
r
s,
”
i
n
2
0
2
4
I
n
t
e
r
n
a
t
i
o
n
a
l
C
o
n
f
e
re
n
c
e
o
n
C
i
r
c
u
i
t
,
S
y
st
e
m
s
a
n
d
C
o
m
m
u
n
i
c
a
t
i
o
n
,
I
C
C
S
C
2
0
2
4
,
I
EEE,
J
u
n
.
2
0
2
4
,
p
p
.
1
–
5
,
d
o
i
:
1
0
.
1
1
0
9
/
I
C
C
S
C
6
2
0
7
4
.
2
0
2
4
.
1
0
6
1
7
0
3
6
.
[
2
5
]
K
.
A
.
S
h
e
e
l
a
a
n
d
K
.
S
.
P
r
a
sa
d
,
“
Li
n
e
a
r
D
i
scri
m
i
n
a
n
t
A
n
a
l
y
s
i
s
F
-
R
a
t
i
o
f
o
r
O
p
t
i
mi
z
a
t
i
o
n
o
f
TESP
A
R
&
M
F
C
C
F
e
a
t
u
r
e
s
f
o
r
S
p
e
a
k
e
r
R
e
c
o
g
n
i
t
i
o
n
,
”
J
o
u
rn
a
l
o
f
Mu
l
t
i
m
e
d
i
a
,
v
o
l
.
2
,
n
o
.
6
,
N
o
v
.
2
0
0
7
,
d
o
i
:
1
0
.
4
3
0
4
/
j
m
m.
2
.
6
.
3
4
-
4
3
.
[
2
6
]
H
.
A
l
i
,
N
.
A
h
m
a
d
,
X
.
Z
h
o
u
,
M
.
A
l
i
,
a
n
d
A
.
A
.
M
a
n
j
o
t
h
o
,
“
L
i
n
e
a
r
d
i
scr
i
mi
n
a
n
t
a
n
a
l
y
si
s
b
a
se
d
a
p
p
r
o
a
c
h
f
o
r
a
u
t
o
m
a
t
i
c
s
p
e
e
c
h
r
e
c
o
g
n
i
t
i
o
n
o
f
u
r
d
u
i
s
o
l
a
t
e
d
w
o
r
d
s
,
”
i
n
C
o
m
m
u
n
i
c
a
t
i
o
n
s
i
n
C
o
m
p
u
t
e
r
a
n
d
I
n
f
o
rm
a
t
i
o
n
S
c
i
e
n
c
e
,
2
0
1
3
,
p
p
.
2
4
–
34
,
d
o
i
:
1
0
.
1
0
0
7
/
9
7
8
-
3
-
3
1
9
-
1
0
9
8
7
-
9
_
3
.
[
2
7
]
N
.
Z
h
e
n
g
,
X
.
W
u
,
J
.
Z
h
o
n
g
,
X
.
L
i
u
,
a
n
d
H
.
M
e
n
g
,
“
S
p
e
a
k
e
r
-
a
w
a
r
e
l
i
n
e
a
r
d
i
s
c
r
i
mi
n
a
n
t
a
n
a
l
y
si
s
i
n
s
p
e
a
k
e
r
v
e
r
i
f
i
c
a
t
i
o
n
,
”
i
n
Pro
c
e
e
d
i
n
g
s
o
f
t
h
e
A
n
n
u
a
l
C
o
n
f
e
re
n
c
e
o
f
t
h
e
I
n
t
e
r
n
a
t
i
o
n
a
l
S
p
e
e
c
h
C
o
m
m
u
n
i
c
a
t
i
o
n
Ass
o
c
i
a
t
i
o
n
,
I
N
T
ERS
PEE
C
H
,
I
S
C
A
:
I
S
C
A
,
O
c
t
.
2
0
2
0
,
p
p
.
3
0
1
2
–
3
0
1
6
,
d
o
i
:
1
0
.
2
1
4
3
7
/
I
n
t
e
r
sp
e
e
c
h
.
2
0
2
0
-
2
0
6
1
.
B
I
O
G
RAP
H
I
E
S O
F
AUTH
O
RS
Abd
e
lm
a
ji
d
La
m
k
a
d
a
m
h
o
l
d
e
r
o
f
a
P
o
stg
ra
d
u
a
te
d
e
g
re
e
in
S
o
ftwa
re
En
g
i
n
e
e
rin
g
a
n
d
a
n
M
BA
i
n
C
o
m
p
u
ter
Ne
two
r
k
s
fr
o
m
t
h
e
F
a
c
u
l
t
y
o
f
S
c
ien
c
e
s.
He
is
a
P
h
.
D.
stu
d
e
n
t
a
t
t
h
e
F
a
c
u
lt
y
o
f
S
c
ien
c
e
s,
F
e
z
,
M
o
ro
c
c
o
.
M
e
m
b
e
r
o
f
L
a
b
o
ra
to
r
y
o
f
E
n
g
in
e
e
rin
g
,
M
o
d
e
li
n
g
a
n
d
An
a
ly
sis
S
y
ste
m
s
(LE
M
S
A).
Cu
rre
n
tl
y
,
h
e
is
a
tea
c
h
e
r
sp
e
c
ializin
g
in
d
ig
it
a
l
in
fra
stru
c
tu
re
(DI)
a
t
th
e
Hig
h
e
r
In
stit
u
te
o
f
A
p
p
li
e
d
Tec
h
n
o
lo
g
y
(HIA
T).
He
c
a
n
b
e
c
o
n
tac
ted
a
t
e
m
a
il
:
a
b
d
e
lma
ji
d
.
la
m
k
a
d
a
m
@u
sm
b
a
.
a
c
.
m
a
.
Mo
h
a
m
m
e
d
K
a
r
im
h
e
i
s
a
P
r
o
f
e
s
s
o
r
o
f
h
i
g
h
e
r
E
d
u
c
a
t
i
o
n
i
n
Co
m
p
u
t
e
r
S
c
i
e
n
c
e
a
n
d
E
l
e
c
t
r
o
n
i
c
s
a
t
t
h
e
F
a
c
u
l
t
y
o
f
S
c
i
e
n
c
e
s
,
F
e
z
,
M
o
r
o
c
c
o
.
H
e
is
t
h
e
I
S
A
I
M
a
s
t
e
r
a
n
d
t
h
e
H
e
a
d
o
f
t
h
e
S
T
I
C
R
e
s
e
a
r
c
h
G
r
o
u
p
.
H
e
s
e
r
v
e
s
a
s
C
h
a
i
rm
a
n
o
f
t
h
e
O
r
g
a
n
i
z
a
t
i
o
n
C
o
m
m
i
t
t
e
e
f
o
r
s
e
v
e
r
a
l
c
o
n
f
e
r
e
n
c
e
s
a
n
d
w
o
r
k
s
h
o
p
s
.
H
e
w
o
r
k
s
o
n
a
v
a
r
ie
t
y
o
f
t
o
p
i
c
s
,
i
n
c
lu
d
i
n
g
e
-
l
e
a
r
n
i
n
g
,
b
i
o
m
e
d
i
c
a
l
e
n
g
i
n
e
e
r
i
n
g
,
a
n
d
c
o
m
p
u
t
e
r
v
i
s
i
o
n
.
M
e
m
b
e
r
o
f
L
a
b
o
r
a
t
o
r
y
o
f
En
g
i
n
e
e
r
i
n
g
,
M
o
d
e
l
i
n
g
a
n
d
A
n
a
l
y
s
i
s
S
y
s
t
e
m
s
(
L
E
M
S
A
)
.
C
u
r
re
n
t
l
y
,
h
e
i
s
t
h
e
d
i
r
e
c
t
o
r
o
f
t
h
e
H
i
g
h
e
r
S
c
h
o
o
l
o
f
T
e
c
h
n
o
l
o
g
y
i
n
F
e
z
,
M
o
r
o
c
c
o
.
H
e
c
a
n
b
e
c
o
n
t
a
c
ted
a
t
e
m
a
i
l
:
m
o
h
a
m
m
e
d
.
k
a
r
i
m
@
u
sm
b
a
.
a
c
.
m
a
.
Evaluation Warning : The document was created with Spire.PDF for Python.