English

Robust Audio-Text Retrieval via Cross-Modal Attention and Hybrid Loss

Computation and Language 2026-04-28 v1 Sound

Abstract

Audio-text retrieval enables semantic alignment between audio content and natural language queries, supporting applications in multimedia search, accessibility, and surveillance. However, current state-of-the-art approaches struggle with long, noisy, and weakly labeled audio due to their reliance on contrastive learning and large-batch training. We propose a novel multimodal retrieval framework that refines audio and text embeddings using a cross-modal embedding refinement module combining transformer-based projection, linear mapping, and bidirectional attention. To further improve robustness, we introduce a hybrid loss function blending cosine similarity, L1\mathcal{L}_{1}, and contrastive objectives, enabling stable training even under small-batch constraints. Our approach efficiently handles long-form and noisy audio (SNR 5 to 15) via silence-aware chunking and attention-based pooling. Experiments on benchmark datasets demonstrate improvements over prior methods.

Keywords

Cite

@article{arxiv.2604.23323,
  title  = {Robust Audio-Text Retrieval via Cross-Modal Attention and Hybrid Loss},
  author = {Meizhu Liu and Matthew Rowe and Amit Agarwal and Michael Avendi and Yassi Abbasi and Hitesh Laxmichand Patel and Paul Li and Kyu J. Han and Tao Sheng and Sujith Ravi and Dan Roth},
  journal= {arXiv preprint arXiv:2604.23323},
  year   = {2026}
}