Correctness consistency¶
CNN14 |
w2v2-b |
hubert-b |
axlstm |
|
|---|---|---|---|---|
Overall Score |
48.6% passed tests (18 passed / 19 failed). |
56.8% passed tests (21 passed / 16 failed). |
64.9% passed tests (24 passed / 13 failed). |
54.1% passed tests (20 passed / 17 failed). |
Samples In Expected High Range¶
Proportion of samples whose predictions fall into the expected value range of >= 0.55
Data |
anger |
fear |
surprise |
|||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|
CNN14 |
w2v2-b |
hubert-b |
axlstm |
CNN14 |
w2v2-b |
hubert-b |
axlstm |
CNN14 |
w2v2-b |
hubert-b |
axlstm |
|
crema-d-1.2.0-emotion.categories.test.gold_standard |
0.74 |
0.75 |
0.79 |
0.70 |
0.42 |
0.40 |
0.40 |
0.43 |
||||
danish-emotional-speech-1.1.1-emotion.test |
0.33 |
0.48 |
0.33 |
0.38 |
0.25 |
0.54 |
0.27 |
0.19 |
||||
emodb-1.2.0-emotion.categories.test.gold_standard |
1.00 |
1.00 |
1.00 |
1.00 |
1.00 |
0.88 |
0.79 |
0.91 |
||||
emovo-1.2.1-emotion.test |
0.86 |
0.90 |
0.89 |
0.94 |
0.46 |
0.54 |
0.42 |
0.40 |
0.46 |
0.67 |
0.58 |
0.60 |
iemocap-2.3.0-emotion.categories.test.gold_standard |
0.73 |
0.83 |
0.82 |
0.75 |
0.41 |
0.65 |
0.65 |
0.41 |
||||
meld-1.3.1-emotion.categories.test.gold_standard |
0.89 |
0.96 |
0.96 |
0.89 |
0.68 |
0.90 |
0.94 |
0.74 |
0.81 |
0.85 |
0.85 |
0.78 |
polish-emotional-speech-1.1.1-emotion.categories.test.gold_standard |
0.92 |
0.92 |
0.95 |
0.88 |
0.32 |
0.45 |
0.42 |
0.30 |
||||
ravdess-1.1.2-emotion.speech.test |
0.88 |
0.88 |
0.97 |
0.91 |
0.59 |
0.50 |
0.44 |
0.72 |
1.00 |
1.00 |
1.00 |
1.00 |
mean |
0.79 |
0.84 |
0.84 |
0.81 |
0.55 |
0.62 |
0.58 |
0.56 |
0.63 |
0.77 |
0.68 |
0.64 |
Samples In Expected Low Range¶
Proportion of samples whose predictions fall into the expected value range of <= 0.45
Data |
boredom |
sadness |
||||||
|---|---|---|---|---|---|---|---|---|
CNN14 |
w2v2-b |
hubert-b |
axlstm |
CNN14 |
w2v2-b |
hubert-b |
axlstm |
|
crema-d-1.2.0-emotion.categories.test.gold_standard |
0.87 |
0.89 |
0.87 |
0.51 |
||||
danish-emotional-speech-1.1.1-emotion.test |
1.00 |
1.00 |
1.00 |
0.88 |
||||
emodb-1.2.0-emotion.categories.test.gold_standard |
0.44 |
0.97 |
0.92 |
0.69 |
0.96 |
1.00 |
1.00 |
1.00 |
emovo-1.2.1-emotion.test |
0.92 |
0.90 |
0.93 |
0.76 |
||||
iemocap-2.3.0-emotion.categories.test.gold_standard |
0.44 |
0.86 |
0.84 |
0.66 |
||||
meld-1.3.1-emotion.categories.test.gold_standard |
0.23 |
0.25 |
0.19 |
0.22 |
||||
polish-emotional-speech-1.1.1-emotion.categories.test.gold_standard |
0.92 |
0.98 |
0.95 |
0.92 |
1.00 |
1.00 |
0.98 |
0.98 |
ravdess-1.1.2-emotion.speech.test |
0.88 |
0.88 |
0.84 |
0.62 |
||||
mean |
0.68 |
0.97 |
0.94 |
0.80 |
0.79 |
0.85 |
0.83 |
0.70 |
Samples In Expected Neutral Range¶
Proportion of samples whose predictions fall into the expected value range of [0.3, 0.6]
Data |
neutral |
|||
|---|---|---|---|---|
CNN14 |
w2v2-b |
hubert-b |
axlstm |
|
crema-d-1.2.0-emotion.categories.test.gold_standard |
0.40 |
0.39 |
0.61 |
0.91 |
danish-emotional-speech-1.1.1-emotion.test |
0.27 |
0.27 |
0.37 |
0.56 |
emodb-1.2.0-emotion.categories.test.gold_standard |
1.00 |
0.52 |
0.78 |
0.93 |
emovo-1.2.1-emotion.test |
0.68 |
0.81 |
0.82 |
0.85 |
iemocap-2.3.0-emotion.categories.test.gold_standard |
0.86 |
0.69 |
0.78 |
0.93 |
meld-1.3.1-emotion.categories.test.gold_standard |
0.73 |
0.48 |
0.45 |
0.76 |
polish-emotional-speech-1.1.1-emotion.categories.test.gold_standard |
0.78 |
0.80 |
0.92 |
0.82 |
ravdess-1.1.2-emotion.speech.test |
0.50 |
0.00 |
0.06 |
0.88 |
mean |
0.65 |
0.49 |
0.60 |
0.83 |
Visualization¶
Distribution of dimensional model predictions for samples with different categorical emotions. The expected range of model predictions is highlighted by the green brackground.
CNN14 |
w2v2-b |
hubert-b |
axlstm |
|---|---|---|---|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|