Skip to main content
. Author manuscript; available in PMC: 2022 Apr 30.
Published in final edited form as: Interspeech. 2021 Aug-Sep;2021:3830–3834. doi: 10.21437/interspeech.2021-2002

Table 4:

Summary of results for AD classification on test set

Feature Set Accu. F1 Spec.
g-LIWC+eGeMAPS 64.79 72.22 61.54
g-BERTsent+Paraling. 74.65 80.56 72.73
Audio Maj. 67.61 77.78 63.49
Top-10 Maj. 80.28 88.89 78.13
Top-10 Avg. 81.69 88.89 80.00
Challenge baseline[6] 78.87 77.78 78.87