Robustness recording condition

Overall scores

w2v2-L-cat

hubert-L-cat

wavlm-cat

data2vec-cat

Overall Score

0.0% passed tests (0 passed / 2 failed).

100.0% passed tests (2 passed / 0 failed).

100.0% passed tests (2 passed / 0 failed).

50.0% passed tests (1 passed / 1 failed).

Percentage Unchanged Predictions Recording Condition

Threshold: 0.8

Data

Percentage Unchanged Predictions Recording Condition

w2v2-L-cat

hubert-L-cat

wavlm-cat

data2vec-cat

imda-nsc-read-speech-balanced-2.6.0-headset-boundary

0.79

0.93

0.85

0.86

imda-nsc-read-speech-balanced-2.6.0-headset-mobile

0.74

0.91

0.83

0.80

mean

0.77

0.92

0.84

0.83

Visualization

Confusion Matrix showing the shift from the predictions of the baseline recording condition audio to the predictions of the different recording condition audio.

w2v2-L-cat

hubert-L-cat

wavlm-cat

data2vec-cat

../../../_images/visualization_imda-nsc-read-speech-balanced-2.6.0-headset-boundary23.png
../../../_images/visualization_imda-nsc-read-speech-balanced-2.6.0-headset-boundary30.png
../../../_images/visualization_imda-nsc-read-speech-balanced-2.6.0-headset-boundary31.png
../../../_images/visualization_imda-nsc-read-speech-balanced-2.6.0-headset-boundary32.png
../../../_images/visualization_imda-nsc-read-speech-balanced-2.6.0-headset-mobile23.png
../../../_images/visualization_imda-nsc-read-speech-balanced-2.6.0-headset-mobile30.png
../../../_images/visualization_imda-nsc-read-speech-balanced-2.6.0-headset-mobile31.png
../../../_images/visualization_imda-nsc-read-speech-balanced-2.6.0-headset-mobile32.png