FPT-Former: A Flexible Parallel Transformer of Recognizing Depression by Using Audiovisual Expert-Knowledge-Based Multimodal Measures

<table class="table-group" id="tab5"><tr><td><table class="table"><tr><td class="thead-hr" colspan="3"><hr/></td></tr><tr class="thead"><td class="align_left">Model variant</td><td class="align_center">RMSE</td><td class="align_center">MAE</td></tr><tr><td class="thead-hr" colspan="3"><hr/></td></tr><tr><td class="align_left">FACS only</td><td class="align_center">6.11</td><td class="align_center">5.91</td></tr><tr><td class="align_left">MFCC only</td><td class="align_center">7.02</td><td class="align_center">6.84</td></tr><tr><td class="align_left">eGeMAPS only</td><td class="align_center">6.60</td><td class="align_center">6.21</td></tr><tr class="table-tr"><td colspan="3"><hr class="tbody-hr"/></td></tr></table></td></tr></table>

<div>The performance of FPT-Former when only one modality is used.</div>

International Journal of Intelligent Systems

tab5

Table 5

Table 5: FPT-Former: A Flexible Parallel Transformer of Recognizing Depression by Using Audiovisual Expert-Knowledge-Based Multimodal Measures 