Robustness low quality phone

Overall scores

	w2v2-b-cat	w2v2-L-cat	w2v2-L-robust-cat	w2v2-L-vox-cat	w2v2-L-xls-r-cat
Overall Score	100.0% passed tests (10 passed / 0 failed).	90.0% passed tests (9 passed / 1 failed).	80.0% passed tests (8 passed / 2 failed).	90.0% passed tests (9 passed / 1 failed).	100.0% passed tests (10 passed / 0 failed).

Change Uar Low Quality Phone

Threshold: -0.05

Data	Change UAR Low Quality Phone
	w2v2-b-cat	w2v2-L-cat	w2v2-L-robust-cat	w2v2-L-vox-cat	w2v2-L-xls-r-cat
crema-d-1.2.0-emotion.categories.test.gold_standard	-0.04	-0.10	-0.13	-0.04	-0.03
emovo-1.2.1-emotion.test	-0.04	0.01	-0.06	-0.07	-0.01
iemocap-2.3.0-emotion.categories.test.gold_standard	-0.03	-0.02	-0.02	-0.03	-0.02
meld-1.3.1-emotion.categories.test.gold_standard	-0.01	-0.01	-0.01	-0.00	0.02
msppodcast-2.6.0-emotion.categories.test-1.gold_standard	-0.04	-0.02	-0.03	-0.02	-0.03
mean	-0.03	-0.03	-0.05	-0.03	-0.01

Percentage Unchanged Predictions Low Quality Phone

Threshold: 0.5

Data	Percentage Unchanged Predictions Low Quality Phone
	w2v2-b-cat	w2v2-L-cat	w2v2-L-robust-cat	w2v2-L-vox-cat	w2v2-L-xls-r-cat
crema-d-1.2.0-emotion.categories.test.gold_standard	0.70	0.71	0.64	0.94	0.87
emovo-1.2.1-emotion.test	0.67	0.72	0.79	0.76	0.82
iemocap-2.3.0-emotion.categories.test.gold_standard	0.71	0.77	0.75	0.77	0.84
meld-1.3.1-emotion.categories.test.gold_standard	0.66	0.74	0.75	0.80	0.67
msppodcast-2.6.0-emotion.categories.test-1.gold_standard	0.68	0.70	0.78	0.74	0.73
mean	0.68	0.73	0.74	0.80	0.79

Visualization

Confusion Matrix showing the shift from the predictions of the original audio to the predictions of the low quality phone audio.

w2v2-b-cat	w2v2-L-cat	w2v2-L-robust-cat	w2v2-L-vox-cat	w2v2-L-xls-r-cat