Robustness small changes

Overall scores

CNN14

w2v2-b

hubert-b

axlstm

Overall Score

70.0% passed tests (14 passed / 6 failed).

75.0% passed tests (15 passed / 5 failed).

60.0% passed tests (12 passed / 8 failed).

35.0% passed tests (7 passed / 13 failed).

Percentage Unchanged Predictions Additive Tone

Threshold: 0.95

Data

Percent Unchanged Pred Additive Tone

CNN14

w2v2-b

hubert-b

axlstm

iemocap-2.3.0-emotion.dimensions.test.gold_standard

0.83

0.92

0.89

0.69

msppodcast-2.6.1-emotion.dimensions.test-1.gold_standard

0.78

0.96

0.94

0.75

mean

0.80

0.94

0.92

0.72

Percentage Unchanged Predictions Append Zeros

Threshold: 0.95

Data

Percent Unchanged Pred Append Zeros

CNN14

w2v2-b

hubert-b

axlstm

iemocap-2.3.0-emotion.dimensions.test.gold_standard

0.99

0.99

0.98

0.83

msppodcast-2.6.1-emotion.dimensions.test-1.gold_standard

1.00

1.00

1.00

0.87

mean

0.99

0.99

0.99

0.85

Percentage Unchanged Predictions Clip

Threshold: 0.95

Data

Percent Unchanged Pred Clip

CNN14

w2v2-b

hubert-b

axlstm

iemocap-2.3.0-emotion.dimensions.test.gold_standard

0.99

0.98

0.97

0.91

msppodcast-2.6.1-emotion.dimensions.test-1.gold_standard

0.99

1.00

1.00

0.98

mean

0.99

0.99

0.98

0.95

Percentage Unchanged Predictions Crop Beginning

Threshold: 0.95

Data

Percent Unchanged Pred Crop Beginning

CNN14

w2v2-b

hubert-b

axlstm

iemocap-2.3.0-emotion.dimensions.test.gold_standard

0.98

0.92

0.91

0.58

msppodcast-2.6.1-emotion.dimensions.test-1.gold_standard

0.99

0.97

0.95

0.60

mean

0.98

0.95

0.93

0.59

Percentage Unchanged Predictions Crop End

Threshold: 0.95

Data

Percent Unchanged Pred Crop End

CNN14

w2v2-b

hubert-b

axlstm

iemocap-2.3.0-emotion.dimensions.test.gold_standard

0.99

0.99

0.99

0.93

msppodcast-2.6.1-emotion.dimensions.test-1.gold_standard

1.00

1.00

0.99

0.97

mean

0.99

0.99

0.99

0.95

Percentage Unchanged Predictions Gain

Threshold: 0.95

Data

Percent Unchanged Pred Gain

CNN14

w2v2-b

hubert-b

axlstm

iemocap-2.3.0-emotion.dimensions.test.gold_standard

0.97

1.00

1.00

1.00

msppodcast-2.6.1-emotion.dimensions.test-1.gold_standard

0.92

1.00

1.00

1.00

mean

0.95

1.00

1.00

1.00

Percentage Unchanged Predictions Highpass Filter

Threshold: 0.95

Data

Percent Unchanged Pred Highpass Filter

CNN14

w2v2-b

hubert-b

axlstm

iemocap-2.3.0-emotion.dimensions.test.gold_standard

0.98

0.99

0.99

0.98

msppodcast-2.6.1-emotion.dimensions.test-1.gold_standard

0.98

0.99

1.00

1.00

mean

0.98

0.99

0.99

0.99

Percentage Unchanged Predictions Lowpass Filter

Threshold: 0.95

Data

Percent Unchanged Pred Lowpass Filter

CNN14

w2v2-b

hubert-b

axlstm

iemocap-2.3.0-emotion.dimensions.test.gold_standard

0.99

1.00

1.00

1.00

msppodcast-2.6.1-emotion.dimensions.test-1.gold_standard

0.81

1.00

1.00

0.94

mean

0.90

1.00

1.00

0.97

Percentage Unchanged Predictions Prepend Zeros

Threshold: 0.95

Data

Percent Unchanged Pred Prepend Zeros

CNN14

w2v2-b

hubert-b

axlstm

iemocap-2.3.0-emotion.dimensions.test.gold_standard

0.98

0.92

0.89

0.55

msppodcast-2.6.1-emotion.dimensions.test-1.gold_standard

0.99

0.97

0.95

0.59

mean

0.98

0.95

0.92

0.57

Percentage Unchanged Predictions White Noise

Threshold: 0.95

Data

Percent Unchanged Pred White Noise

CNN14

w2v2-b

hubert-b

axlstm

iemocap-2.3.0-emotion.dimensions.test.gold_standard

0.52

0.83

0.74

0.65

msppodcast-2.6.1-emotion.dimensions.test-1.gold_standard

0.61

0.83

0.87

0.72

mean

0.56

0.83

0.80

0.69

Visualization

CNN14

w2v2-b

hubert-b

axlstm

../../../_images/visualization_iemocap-2.3.0-emotion.dimensions.test.gold_standard100.png
../../../_images/visualization_iemocap-2.3.0-emotion.dimensions.test.gold_standard101.png
../../../_images/visualization_iemocap-2.3.0-emotion.dimensions.test.gold_standard102.png
../../../_images/visualization_iemocap-2.3.0-emotion.dimensions.test.gold_standard103.png
../../../_images/visualization_msppodcast-2.6.1-emotion.dimensions.test-1.gold_standard152.png
../../../_images/visualization_msppodcast-2.6.1-emotion.dimensions.test-1.gold_standard153.png
../../../_images/visualization_msppodcast-2.6.1-emotion.dimensions.test-1.gold_standard154.png
../../../_images/visualization_msppodcast-2.6.1-emotion.dimensions.test-1.gold_standard155.png