Skip to main content
. Author manuscript; available in PMC: 2022 Nov 4.
Published in final edited form as: Interspeech. 2022 Sep;2022:3338–3342. doi: 10.21437/interspeech.2022-10798

Table 3:

Percentage change (relative) in F1-Avg after adversarial disentanglement for three input features using DAIC-WOZ and CONVERGE datasets.

Dataset Input Feature % Change in F1-Avg
DIAC-WOZ Mel-spec +4.36
Raw-audio +2.16
Wav2vec2.0 +0.87
CONVERGE Mel-spec +1.25
Raw-audio +3.34
XLSR-Mandarin +0.33*
*

indicates that the change is not statistically significant.