SHARED MULTI-MODAL EMBEDDING SPACE FOR FACE-VOICE ASSOCIATION

Christopher Simic

ICASSP 2026poster0 citations

SHARED MULTI-MODAL EMBEDDING SPACE FOR FACE-VOICE ASSOCIATION

Abstract

The FAME 2026 challenge comprises two demanding tasks: training face-voice associations combined with a multilingual setting that includes testing on languages on which the model was not trained. Our approach consists of separate uni-modal processing pipelines with general face and voice feature extraction, complemented by additional age-gender feature extraction to support prediction. The resulting single-modal features are projected into a shared embedding space and trained with an Adaptive Angular Margin (AAM) loss. Our approach achieved first place in the FAME 2026 challenge, with an average Equal-Error Rate (EER) of 23.99%.

BibTeX

@inproceedings{icassp2026_sharedmultimodal,
  title = {SHARED MULTI-MODAL EMBEDDING SPACE FOR FACE-VOICE ASSOCIATION},
  author = {Christopher Simic},
  booktitle = {ICASSP 2026},
  year = {2026}
}