# CLAPSep: Leveraging Contrastive Pre-trained Model for Multi-Modal Query-Conditioned Target Sound Extraction

arXiv: https://arxiv.org/abs/2402.17455 · HTML: https://mlsanity.com/paper/2402.17455/

## Key facts

- Cited by 9 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 27 Feb 2024.

## Built on by (most cited first, 9 of 9)

- [MGE-LDM: Joint Latent Diffusion for Simultaneous Music Generation and Source Extraction](https://mlsanity.com/paper/2505.23305/) (2025)
- [Unsupervised Single-Channel Audio Separation with Diffusion Source Priors](https://mlsanity.com/paper/2512.07226/) (2025)
- [Multichannel-to-Multichannel Target Sound Extraction Using Direction and Timestamp Clues](https://mlsanity.com/paper/2409.12415/) (2024)
- [MARS-Sep: Multimodal-Aligned Reinforced Sound Separation](https://mlsanity.com/paper/2510.10509/) (2025)
- [SAM Audio Judge: A Unified Multimodal Framework for Perceptual Evaluation of Audio Separation](https://mlsanity.com/paper/2601.19702/) (2026)
- [SelectTSL: Prompt-Guided Selective Target Sound Localization in Complex Scenarios](https://mlsanity.com/paper/2607.02343/) (2026)
- [SoundCompass: Navigating Target Sound Extraction With Effective Directional Clue Integration In Complex Acoustic Scenes](https://mlsanity.com/paper/2509.18561/) (2025)
- [Contour-Guided Query-Based Feature Fusion for Boundary-Aware and Generalizable Cardiac Ultrasound Segmentation](https://mlsanity.com/paper/2603.28110/) (2026)
- [Long-Term Memory-Guided Enhancement for Target Perception in Audio-Language Models](https://mlsanity.com/paper/2609.36577/) (2026)

## Abstract (arXiv)

Universal sound separation (USS) aims to extract arbitrary types of sounds from real-world recordings. This can be achieved by language-queried target sound extraction (TSE), which typically consists of two components: a query network that converts user queries into conditional embeddings, and a separation network that extracts the target sound accordingly. Existing methods commonly train models from scratch. As a consequence, substantial data and computational resources are required to make the randomly initialized model comprehend sound events and perform separation accordingly. In this paper, we propose to integrate pre-trained models into TSE models to address the above issue. To be specific, we tailor and adapt the powerful contrastive language-audio pre-trained model (CLAP) for USS, denoted as CLAPSep. CLAPSep also accepts flexible user inputs, taking both positive and negative user prompts of uni- and/or multi-modalities for target sound extraction. These key features of CLAPSep can not only enhance the extraction performance but also improve the versatility of its application. We provide extensive experiments on 5 diverse datasets to demonstrate the superior performance and zero- and few-shot generalizability of our proposed CLAPSep with fast training convergence, surpassing previous methods by a significant margin. Full codes and some audio examples are released for reproduction and evaluation.