license: [cc-by-nc-sa-4.0](https://creativecommons.org/licenses/by-nc-sa/4.0/)

Example audios are from [TestableFred](https://space.bilibili.com/258080618)'s video: https://www.bilibili.com/video/BV1UZUpYGEwM

Before inputting it into the model for inference, the original audio was separated from the vocal and instrument using the `model_bs_roformer_ep_368_sdr_12.9628.ckpt` model.