Skip to main content

Table 2 Confusion matrix for HMM-based system using glottal parameter contours (S-I), ν i = s F g i , b F g i , s A g i , b A g i (overall accuracy 52.6%)

From: On the use of speech parameter contours for emotion recognition

 

Neutral (%)

Anger (%)

Sad (%)

Happy (%)

Bored (%)

Neutral

53.2

0

14.9

6.4

25.5

Anger

1.4

80.3

0

18.3

0

Sad

16.4

3.3

31.2

6.6

42.6

Happy

9.6

19.2

6.9

50.7

13.7

Bored

13.0

1.3

31.2

9.1

45.5