Fully Few-shot Class-incremental Audio Classification with Adaptive Improvement of Stability and Plasticity
This repository contains the introductions to the datasets and codes used in our paper, titled "Fully Few-shot Class-incremental Audio Classification with Adaptive Improvement of Stability and Plasticity " .
Three audio datasets, including FSC-89, NSynth-100 and LS-100, are adopted as experimental datasets to evaluate the performance of different methods, which have been widely used in previous works for audio classification. To facilitate reimplementation of the results of this paper, the details of these three audio datasets are described at three websites . What's more, they can be downloaded from the above three websites and be freely used for research purpose.
https://www.modelscope.cn/datasets/pp199124903/LS-100/summary
https://www.modelscope.cn/datasets/pp199124903/FSC-89/summary
https://www.modelscope.cn/datasets/pp199124903/NSynth-100/summary
Yanxiong Li ([email protected]) and Yongjie Si ([email protected]) School of Electronic and Information Engineering, South China University of Technology, Guangzhou, China