# Coarse-to-fine Alignment Makes Better Speech-image Retrieval

arXiv: https://arxiv.org/abs/2408.13119 · HTML: https://mlsanity.com/paper/2408.13119/

## Key facts

- Cited by 1 papers in the map (data through 14 Oct 2026).
- Builds on 20 of its 25 references that are in the map.
- First posted on arXiv 15 Aug 2024.

## Builds on (most cited first, 20 of 20)

- [BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding](https://mlsanity.com/paper/1810.04805/) (2018)
- [Learning Transferable Visual Models From Natural Language Supervision](https://mlsanity.com/paper/2103.00020/) (2021)
- [Momentum Contrast for Unsupervised Visual Representation Learning](https://mlsanity.com/paper/1911.05722/) (2019)
- [BLIP-2: Bootstrapping Language-Image Pre-training with Frozen Image Encoders and Large Language Models](https://mlsanity.com/paper/2301.12597/) (2023)
- [HuBERT: Self-Supervised Speech Representation Learning by Masked Prediction of Hidden Units](https://mlsanity.com/paper/2106.07447/) (2021)
- [Align before Fuse: Vision and Language Representation Learning with Momentum Distillation](https://mlsanity.com/paper/2107.07651/) (2021)
- [Deep Visual-Semantic Alignments for Generating Image Descriptions](https://mlsanity.com/paper/1412.2306/) (2014)
- [SUPERB: Speech processing Universal PERformance Benchmark](https://mlsanity.com/paper/2105.01051/) (2021)
- [End-to-end ASR: from Supervised to Semi-Supervised Learning with Modern Architectures](https://mlsanity.com/paper/1911.08460/) (2019)
- [Transformer-based Acoustic Modeling for Hybrid Speech Recognition](https://mlsanity.com/paper/1910.09799/) (2019)
- [SUPERB-SG: Enhanced Speech processing Universal PERformance Benchmark for Semantic and Generative Capabilities](https://mlsanity.com/paper/2203.06849/) (2022)
- [Deep Multimodal Semantic Embeddings for Speech and Images](https://mlsanity.com/paper/1511.03690/) (2015)
- [Text-Free Image-to-Speech Synthesis Using Learned Segmental Units](https://mlsanity.com/paper/2012.15454/) (2020)
- [Improving Depth Gradient Continuity in Transformers: A Comparative Study on Monocular Depth Estimation with CNN](https://mlsanity.com/paper/2308.08333/) (2023)
- [SpeechCLIP: Integrating Speech with Pre-Trained Vision and Language Model](https://mlsanity.com/paper/2210.00705/) (2022)
- [Fast-Slow Transformer for Visually Grounding Speech](https://mlsanity.com/paper/2109.08186/) (2021)
- [Multi-Modal Proxy Learning Towards Personalized Visual Multiple Clustering](https://mlsanity.com/paper/2404.15655/) (2024)
- [Talk, Don't Write: A Study of Direct Speech-Based Image Retrieval](https://mlsanity.com/paper/2104.01894/) (2021)
- [Transfer Learning from Audio-Visual Grounding to Speech Recognition](https://mlsanity.com/paper/1907.04355/) (2019)
- [Leveraging Pretrained Image-text Models for Improving Audio-Visual Learning](https://mlsanity.com/paper/2309.04628/) (2023)

## Built on by (most cited first, 1 of 1)

- [Exploring Multi-Modal Large Language Models and Two-Stage Fine-Tuning for Fashion Image Retrieval](https://mlsanity.com/paper/2606.19684/) (2026)

## Abstract (arXiv)

In this paper, we propose a novel framework for speech-image retrieval. We utilize speech-image contrastive (SIC) learning tasks to align speech and image representations at a coarse level and speech-image matching (SIM) learning tasks to further refine the fine-grained cross-modal alignment. SIC and SIM learning tasks are jointly trained in a unified manner. To optimize the learning process, we utilize an embedding queue that facilitates efficient sampling of high-quality and diverse negative representations during SIC learning. Additionally, it enhances the learning of SIM tasks by effectively mining hard negatives based on contrastive similarities calculated in SIC tasks. To further optimize learning under noisy supervision, we incorporate momentum distillation into the training process. Experimental results show that our framework outperforms the state-of-the-art method by more than 4% in R@1 on two benchmark datasets for the speech-image retrieval tasks. Moreover, as observed in zero-shot experiments, our framework demonstrates excellent generalization capabilities.