Abstract
We study the task of learning association between faces and voices, which is gaining interest in the multimodal community lately. These methods suffer from the deliberate crafting of negative mining procedures as well as the reliance on the distant margin parameter. These issues are addressed by learning a joint embedding space in which orthogonality constraints are applied to the fused embeddings of faces and voices. However, embedding spaces of faces and voices possess different characteristics and require spaces to be aligned before fusing them. To this end, we propose a method that accurately aligns the embedding spaces and fuses them with an enhanced gated fusion thereby improving the performance of face-voice association. Extensive experiments on the VoxCeleb dataset reveals the merits of the proposed approach.
Citation
Abdul Hannan,
Muhammad Arslan Manzoor,
Nawaz,
Muhammad Irzam Liaqat,
Schedl,
Mubashir Noman
PAEFF: Precise Alignment and Enhanced Gated Feature Fusion for Face-Voice Association
Proceedings of the Annual Conference of the International Speech Communication Association, 2025.
BibTeX
@inproceedings{AbdulHannan2025l_speech_recog,
title = {PAEFF: Precise Alignment and Enhanced Gated Feature Fusion for Face-Voice Association},
author = {Abdul Hannan and Muhammad Arslan Manzoor and Nawaz and Muhammad Irzam Liaqat and Schedl and Mubashir Noman},
booktitle = {Proceedings of the Annual Conference of the International Speech Communication Association},
year = {2025}
}