# Few-shot learning of new sound classes for target sound extraction

arXiv: https://arxiv.org/abs/2106.07144 · HTML: https://mlsanity.com/paper/2106.07144/

## Key facts

- Cited by 5 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 14 Jun 2021.

## Built on by (most cited first, 5 of 5)

- [Real-Time Target Sound Extraction](https://mlsanity.com/paper/2211.02250/) (2022)
- [DPM-TSE: A Diffusion Probabilistic Model for Target Sound Extraction](https://mlsanity.com/paper/2310.04567/) (2023)
- [SoloAudio: Target Sound Extraction with Language-oriented Audio Diffusion Transformer](https://mlsanity.com/paper/2409.08425/) (2024)
- [Typing to Listen at the Cocktail Party: Text-Guided Target Speaker Extraction](https://mlsanity.com/paper/2310.07284/) (2023)
- [A Comprehensive Survey of Challenges and Opportunities of Few-Shot Learning Across Multiple Domains](https://mlsanity.com/paper/2504.04017/) (2025)

## Abstract (arXiv)

Target sound extraction consists of extracting the sound of a target acoustic event (AE) class from a mixture of AE sounds. It can be realized using a neural network that extracts the target sound conditioned on a 1-hot vector that represents the desired AE class. With this approach, embedding vectors associated with the AE classes are directly optimized for the extraction of sound classes seen during training. However, it is not easy to extend this framework to new AE classes, i.e. unseen during training. Recently, speech, music, or AE sound extraction based on enrollment audio of the desired sound offers the potential of extracting any target sound in a mixture given only a short audio signal of a similar sound. In this work, we propose combining 1-hot- and enrollment-based target sound extraction, allowing optimal performance for seen AE classes and simple extension to new classes. In experiments with synthesized sound mixtures generated with the Freesound Dataset (FSD) datasets, we demonstrate the benefit of the combined framework for both seen and new AE classes. Besides, we also propose adapting the embedding vectors obtained from a few enrollment audio samples (few-shot) to further improve performance on new classes.