Correctness speaker average

Overall scores

w2v2-L

hubert-L

wavlm

data2vec

Overall Score

100.0% passed tests (3 passed / 0 failed).

100.0% passed tests (3 passed / 0 failed).

66.7% passed tests (2 passed / 1 failed).

100.0% passed tests (3 passed / 0 failed).

Mean Absolute Error

Threshold: 0.1

Data

Mean Absolute Error

w2v2-L

hubert-L

wavlm

data2vec

iemocap-2.3.0-full

0.03

0.03

0.09

0.02

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

0.05

0.03

0.03

0.04

msppodcast-2.6.1-emotion.dimensions.test-2.gold_standard

0.04

0.03

0.10

0.04

mean

0.04

0.03

0.07

0.03

Visualization

The plot shows the predicted average value with the true average value. We select a slightly higher threshold for the absolute error in the plots compared to the Mean Absolute Error test as we are interested in highlighting only big deviations here.

w2v2-L

hubert-L

wavlm

data2vec

../../../_images/visualization_iemocap-2.3.0-full4.png
../../../_images/visualization_iemocap-2.3.0-full8.png
../../../_images/visualization_iemocap-2.3.0-full9.png
../../../_images/visualization_iemocap-2.3.0-full10.png
../../../_images/visualization_msppodcast-2.6.1-emotion.dimensions.test-1.gold_standard32.png
../../../_images/visualization_msppodcast-2.6.1-emotion.dimensions.test-1.gold_standard54.png
../../../_images/visualization_msppodcast-2.6.1-emotion.dimensions.test-1.gold_standard55.png
../../../_images/visualization_msppodcast-2.6.1-emotion.dimensions.test-1.gold_standard56.png
../../../_images/visualization_msppodcast-2.6.1-emotion.dimensions.test-2.gold_standard24.png
../../../_images/visualization_msppodcast-2.6.1-emotion.dimensions.test-2.gold_standard38.png
../../../_images/visualization_msppodcast-2.6.1-emotion.dimensions.test-2.gold_standard39.png
../../../_images/visualization_msppodcast-2.6.1-emotion.dimensions.test-2.gold_standard40.png