Shared Multi-Modal Embedding Space for Face-Voice Association
- 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 on 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%.
| Author: | Christopher Simic, Korbinian RiedhammerORCiD, Tobias BockletORCiD |
|---|---|
| DOI: | https://doi.org/10.1109/ICASSP55912.2026.11461404 |
| Parent Title (English): | ICASSP 2026 - 2026 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP) |
| Publisher: | IEEE |
| Document Type: | conference proceeding (article) |
| Language: | English |
| Reviewed: | Begutachtet/Reviewed |
| Release Date: | 2026/06/11 |
| Tag: | Age-gender Prediction; Face Recognition; Multi-Modal; Speaker Recognition |
| Volume: | 2026 |
| Pagenumber: | 3 |
| First Page: | 21766 |
| Last Page: | 21768 |
| institutes: | Fakultät Informatik |
| Zentrum für Künstliche Intelligenz (KIZ) | |
| Research Themes: | Digitalisierung & Künstliche Intelligenz |
