Skip to main content
. Author manuscript; available in PMC: 2022 Nov 1.
Published in final edited form as: Proc IEEE Int Conf Acoust Speech Signal Process. 2022 Apr 27;2022:6267–6271. doi: 10.1109/icassp43922.2022.9746307

Table 4.

Results, in terms of F1-score, for depression detection on the CONVERGE dataset using x-vector embeddings with a CNN classifier as the backend, with and without FrAUG. The best F1-score is boldfaced.

L,R Configuration Validation Test Data
Augmentation
Baseline (L=64ms,R=50%) 0.676 0.654 None
L=32ms, 64ms
R=50%, 25%
0.705 0.712 3x
L=32ms, 64ms
R=50%, 25%, 10%
0.720 0.719 5x
L=32ms, 64ms, 128ms
R=50%, 25%, 10%
0.739 0.739 8x