Skip to main content

Table 5 The P EL and P NR results when morphological image processing is applied

From: Improved monaural speech segregation based on computational auditory scene analysis

Intrusion

Morphological

Hu–Wang tandem

Mixture

image processing

system

 
 

PEL(%)

PNR(%)

PEL(%)

PNR(%)

PNR(%)

N1

2.98

2.08

3.34

2.46

27.11

N2

4.96

5.05

4.76

6.08

29.50

N3

5.81

4.06

6.24

5.62

40.56

N4

2.64

4.91

3.03

5.82

22.01

N5

1.89

1.07

2.39

1.64

9.77

N6

1.98

1.92

4.65

2.17

46.31

N7

5.70

5.28

6.19

6.63

36.51

N8

6.19

0.88

7.07

3.23

18.43

N9

4.32

4.87

7.49

5.21

29.46

N10

2.55

0.56

3.84

0.58

11.77

N11

4.01

2.89

6.38

3.02

20.73

N12

2.05

2.98

4.67

3.71

76.79

N13

4.43

2.44

5.09

3.12

13.76

N14

4.81

3.95

6.03

4.80

23.16

N15

3.79

4.46

4.18

5.78

26.84

N16

1.66

1.94

2.03

2.89

6.64

N17

3.70

6.33

6.37

6.94

43.07

Average

3.73

3.27

4.93

4.10

28.38