Robustness spectral tilt

Overall scores

	w2v2-b	w2v2-L	w2v2-L-robust	w2v2-L-xls-r	w2v2-L-vox
Overall Score	75.0% passed tests (6 passed / 2 failed).	75.0% passed tests (6 passed / 2 failed).	87.5% passed tests (7 passed / 1 failed).	75.0% passed tests (6 passed / 2 failed).	62.5% passed tests (5 passed / 3 failed).

Change Ccc Downward Tilt

Threshold: -0.05

Data	Change CCC Downward Tilt
	w2v2-b	w2v2-L	w2v2-L-robust	w2v2-L-xls-r	w2v2-L-vox
iemocap-2.3.0-emotion.dimensions.test.gold_standard	-0.00	-0.01	-0.01	-0.02	-0.01
msppodcast-2.6.1-emotion.dimensions.test-1.gold_standard	-0.01	-0.00	-0.02	-0.01	-0.02
mean	-0.01	-0.01	-0.01	-0.01	-0.01

Change Ccc Upward Tilt

Threshold: -0.05

Data	Change CCC Upward Tilt
	w2v2-b	w2v2-L	w2v2-L-robust	w2v2-L-xls-r	w2v2-L-vox
iemocap-2.3.0-emotion.dimensions.test.gold_standard	-0.01	0.01	0.01	0.00	-0.03
msppodcast-2.6.1-emotion.dimensions.test-1.gold_standard	-0.01	-0.00	-0.01	-0.01	-0.06
mean	-0.01	0.01	0.00	-0.01	-0.04

Percentage Unchanged Predictions Downward Tilt

Threshold: 0.8

Data	Percentage Unchanged Predictions Downward Tilt
	w2v2-b	w2v2-L	w2v2-L-robust	w2v2-L-xls-r	w2v2-L-vox
iemocap-2.3.0-emotion.dimensions.test.gold_standard	0.95	0.94	0.98	0.95	0.96
msppodcast-2.6.1-emotion.dimensions.test-1.gold_standard	0.92	0.90	0.92	0.86	0.89
mean	0.94	0.92	0.95	0.91	0.93

Percentage Unchanged Predictions Upward Tilt

Threshold: 0.8

Data	Percentage Unchanged Predictions Upward Tilt
	w2v2-b	w2v2-L	w2v2-L-robust	w2v2-L-xls-r	w2v2-L-vox
iemocap-2.3.0-emotion.dimensions.test.gold_standard	0.72	0.59	0.75	0.57	0.62
msppodcast-2.6.1-emotion.dimensions.test-1.gold_standard	0.72	0.72	0.86	0.51	0.57
mean	0.72	0.66	0.80	0.54	0.59

Visualization Downward Tilt

Difference of predictions for original audio and audio with a downward spectral tilt. 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.

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