XM-ALIGN: Unified Cross-Modal Embedding Alignment for Face-Voice Association

Zhihua Fang, Shumei Tao, Junxu Wang, Liang He
12/7/2025
cs.SDcs.CV

Official code: PunkMale/XM-ALIGN (7 stars, MIT, last commit 2026-05-01).

Abstract

This paper introduces our solution, XM-ALIGN (Unified Cross-Modal Embedding Alignment Framework), proposed for the FAME challenge at ICASSP 2026. Our framework combines explicit and implicit alignment mechanisms, significantly improving cross-modal verification performance in both "heard" and "unheard" languages. By extracting feature embeddings from both face and voice encoders and jointly optimizing them using a shared classifier, we employ mean squared error (MSE) as the embedding alignment loss to ensure tight alignment between modalities. Additionally, data augmentation strategies are applied during model training to enhance generalization. Experimental results show that our approach demonstrates superior performance on the MAV-Celeb dataset. The code will be released at https://github.com/PunkMale/XM-ALIGN.

View on arXivView PDF

Code Implementations(1)

PunkMale/XM-ALIGNOfficial100%
MIT

ICASSP'26｜XM-ALIGN: Unified Cross-Modal Embedding Alignment for Face-Voice Association

70PythonNov 21, 20255 months agoMIT

Similar Papers

Shared Multi-modal Embedding Space for Face-Voice Association

85% match2025

Nexus: An Omni-Perceptive And -Interactive Model for Language, Audio, And Vision

83% match1 repo32025

Decoupled Audio-Visual Dataset Distillation

82% match2025

Modest-Align: Data-Efficient Alignment for Vision-Language Models

82% match2025

EgoVIS@CVPR: PAIR-Net: Enhancing Egocentric Speaker Detection via Pretrained Audio-Visual Fusion and Alignment Loss

82% match2025

Visatronic: A Multimodal Decoder-Only Model for Speech Synthesis

81% match1 repo162024

Cite this paper

@article{fang2025xmalign,
  title  = {XM-ALIGN: Unified Cross-Modal Embedding Alignment for Face-Voice Association},
  author = {Zhihua Fang and Shumei Tao and Junxu Wang and Liang He},
  year   = {2025},
  eprint = {2512.06757},
  archivePrefix = {arXiv},
  url    = {http://arxiv.org/abs/2512.06757v1}
}

Add a badge to your README

Shows whether official code is linked for this paper.

Code of Paper badge
[![Code of Paper](https://codeofpaper.com/badge/2512.06757.svg)](https://codeofpaper.com/paper/2512.06757)

Discussion