English

CatchPhrase: EXPrompt-Guided Encoder Adaptation for Audio-to-Image Generation

Multimedia 2025-07-28 v1 Sound Audio and Speech Processing

Abstract

We propose CatchPhrase, a novel audio-to-image generation framework designed to mitigate semantic misalignment between audio inputs and generated images. While recent advances in multi-modal encoders have enabled progress in cross-modal generation, ambiguity stemming from homographs and auditory illusions continues to hinder accurate alignment. To address this issue, CatchPhrase generates enriched cross-modal semantic prompts (EXPrompt Mining) from weak class labels by leveraging large language models (LLMs) and audio captioning models (ACMs). To address both class-level and instance-level misalignment, we apply multi-modal filtering and retrieval to select the most semantically aligned prompt for each audio sample (EXPrompt Selector). A lightweight mapping network is then trained to adapt pre-trained text-to-image generation models to audio input. Extensive experiments on multiple audio classification datasets demonstrate that CatchPhrase improves audio-to-image alignment and consistently enhances generation quality by mitigating semantic misalignment.

Keywords

Cite

@article{arxiv.2507.18750,
  title  = {CatchPhrase: EXPrompt-Guided Encoder Adaptation for Audio-to-Image Generation},
  author = {Hyunwoo Oh and SeungJu Cha and Kwanyoung Lee and Si-Woo Kim and Dong-Jin Kim},
  journal= {arXiv preprint arXiv:2507.18750},
  year   = {2025}
}
R2 v1 2026-07-01T04:17:45.647Z