DSD100曲目数据集

The dsd100 is a dataset of 100 full lengths music tracks of different styles along with their isolated drums, bass, vocals and others stems. dsd100 contains two folders, a folder with a training set: "train", composed of 50 songs, and a folder with a test set: "test", composed of 50 songs. Supervised approaches should be trained on the training set and tested on both sets. For each file, the mixture correspond to the sum of all the signals. All signals are stereophonic and encoded at 44.1kHz. The data from dsd100 consist of 100 tracks which are derived from The 'Mixing Secrets' Free Multitrack Download Library. Please refer to this original resource for any question regarding your rights on your use of the DSD100 data. Have a look at the detailed list of all tracks. We would like to thank Mike Senior not only for giving us the permission to use this multitrack material, but also for maintaining such resources for the audio community. If you use this dataset, please reference it accordingly:

DSD100（dsd100）是一个包含100首不同风格完整音乐曲目及其分离鼓组、贝斯、人声及其他分轨素材（stems）的数据集。该数据集包含两个文件夹：分别为含50首歌曲的训练集文件夹"train"，与含50首歌曲的测试集文件夹"test"。监督学习方法应在训练集上完成模型训练，并在两个数据集上开展测试。对于每个文件，混合音轨对应所有信号的总和。所有信号均为立体声，编码采样率为44.1kHz。DSD100的100首曲目均源自《Mixing Secrets》免费多轨素材下载库。若您对使用DSD100数据的相关权利存在任何疑问，请查阅该原始资源。您可查看所有曲目的详细列表。我们谨向Mike Senior致谢：不仅感谢其允许我们使用这批多轨素材，更感谢其为音频社区维护了如此优质的资源。若您使用本数据集，请按规范进行引用。

搜集汇总

背景与挑战

背景概述

DSD100曲目数据集包含100首不同风格的音乐曲目，每首曲目提供分离的鼓、贝斯、人声和其他音轨，分为50首训练集和50首测试集，音频为44.1kHz立体声。数据集来源于'The Mixing Secrets' Free Multitrack Download Library，并附有解析工具和评估方法。