Correctness speaker average¶
w2v2-L |
hubert-L |
wavlm |
data2vec |
|
|---|---|---|---|---|
Overall Score |
100.0% passed tests (3 passed / 0 failed). |
100.0% passed tests (3 passed / 0 failed). |
100.0% passed tests (3 passed / 0 failed). |
100.0% passed tests (3 passed / 0 failed). |
Mean Absolute Error¶
Data |
Mean Absolute Error |
|||
|---|---|---|---|---|
w2v2-L |
hubert-L |
wavlm |
data2vec |
|
iemocap-2.3.0-full |
0.04 |
0.03 |
0.08 |
0.03 |
msppodcast-2.6.1-emotion.dimensions.test-1.gold_standard |
0.04 |
0.04 |
0.03 |
0.05 |
msppodcast-2.6.1-emotion.dimensions.test-2.gold_standard |
0.03 |
0.03 |
0.08 |
0.04 |
mean |
0.04 |
0.03 |
0.06 |
0.04 |
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 |
|---|---|---|---|
|
|
|
|
|
|
|
|
|
|
|
|