Robustness low quality phone

Overall scores

CNN14

w2v2-b

hubert-b

axlstm

Overall Score

0.0% passed tests (0 passed / 4 failed).

0.0% passed tests (0 passed / 4 failed).

0.0% passed tests (0 passed / 4 failed).

25.0% passed tests (1 passed / 3 failed).

Change Ccc Low Quality Phone

Threshold: -0.05

Data

Change CCC Low Quality Phone

CNN14

w2v2-b

hubert-b

axlstm

iemocap-2.3.0-emotion.dimensions.test.gold_standard

-0.11

-0.10

-0.06

0.06

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

-0.16

-0.24

-0.07

-0.12

mean

-0.14

-0.17

-0.07

-0.03

Percentage Unchanged Predictions Low Quality Phone

Threshold: 0.5

Data

Percentage Unchanged Predictions Low Quality Phone

CNN14

w2v2-b

hubert-b

axlstm

iemocap-2.3.0-emotion.dimensions.test.gold_standard

0.35

0.42

0.43

0.22

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

0.29

0.29

0.45

0.23

mean

0.32

0.35

0.44

0.23

Visualization

Difference of predictions for original audio and low quality phone audio. The allowed prediction difference \(\delta < 0.05\) is highlighted in green in the upper plot. The lower plot shows the distributions of the two predictions.

CNN14

w2v2-b

hubert-b

axlstm

../../../_images/visualization_iemocap-2.3.0-emotion.dimensions.test.gold_standard96.png
../../../_images/visualization_iemocap-2.3.0-emotion.dimensions.test.gold_standard97.png
../../../_images/visualization_iemocap-2.3.0-emotion.dimensions.test.gold_standard98.png
../../../_images/visualization_iemocap-2.3.0-emotion.dimensions.test.gold_standard99.png
../../../_images/visualization_msppodcast-2.6.1-emotion.dimensions.test-1.gold_standard148.png
../../../_images/visualization_msppodcast-2.6.1-emotion.dimensions.test-1.gold_standard149.png
../../../_images/visualization_msppodcast-2.6.1-emotion.dimensions.test-1.gold_standard150.png
../../../_images/visualization_msppodcast-2.6.1-emotion.dimensions.test-1.gold_standard151.png