free-music-archive-large

# FMA: A Dataset for Music Analysis

[Michaël Defferrard](https://deff.ch/), [Kirell Benzi](https://kirellbenzi.com/), [Pierre Vandergheynst](https://people.epfl.ch/pierre.vandergheynst), [Xavier Bresson](https://www.ntu.edu.sg/home/xbresson).

**International Society for Music Information Retrieval Conference (ISMIR), 2017.**

> We introduce the Free Music Archive (FMA), an open and easily accessible dataset suitable for evaluating several tasks in MIR, a field concerned with browsing, searching, and organizing large music collections. The community's growing interest in feature and end-to-end learning is however restrained by the limited availability of large audio datasets. The FMA aims to overcome this hurdle by providing 917 GiB and 343 days of Creative Commons-licensed audio from 106,574 tracks from 16,341 artists and 14,854 albums, arranged in a hierarchical taxonomy of 161 genres. It provides full-length and high-quality audio, pre-computed features, together with track- and user-level metadata, tags, and free-form text such as biographies. We here describe the dataset and how it was created, propose a train/validation/test split and three subsets, discuss some suitable MIR tasks, and evaluate some baselines for genre recognition. Code, data, and usage examples are available at https://github.com/mdeff/fma.

Paper: [arXiv:1612.01840](https://arxiv.org/abs/1612.01840) - [latex and reviews](https://github.com/mdeff/paper-fma-ismir2017)

Slides: [doi:10.5281/zenodo.1066119](https://doi.org/10.5281/zenodo.1066119)

Poster: [doi:10.5281/zenodo.1035847](https://doi.org/10.5281/zenodo.1035847)

# This Pack

This is the **large** dataset, comprising a total of **105,024 samples** clipped at **30 seconds** over **16** *unbalanced* genres totaling **869.2 hours** hours of audio.

## Repack Notes

- 173 files were unreadable by `libsndfile / libmpg123`, these were removed.
- 1377 files had licenses that were unclear on whether or not they permitted redistribution, or the full license text was unavailable. These were removed.

# License

- The [FMA codebase](https://github.com/mdeff/fma) is released under [The MIT License](https://github.com/mdeff/fma/blob/master/LICENSE.txt).
- The FMA metadata is released under [CC-BY 4.0](https://creativecommons.org/licenses/by/4.0).
- The individual files are released under various Creative Commons family licenses, with a small amount of additional licenses. **Each file has its license attached and important details of the license enumerated.** To make it easy to use for developers and trainers, a configuration is available to limit only to commercially-usable data.

Please refer to any of the following URLs for additional details.

| Class Label | License Name | URL |
| ----------- | ------------ | --- |
| 0 | CC-BY 1.0 | https://creativecommons.org/licenses/by/1.0/ |
| 1 | CC-BY 2.0 | https://creativecommons.org/licenses/by/2.0/ |
| 2 | CC-BY 2.5 | https://creativecommons.org/licenses/by/2.5/ |
| 3 | CC-BY 3.0 | https://creativecommons.org/licenses/by/3.0/ |
| 4 | CC-BY 4.0 | https://creativecommons.org/licenses/by/4.0/ |
| 5 | CC-BY-NC 2.0 | https://creativecommons.org/licenses/by-nc/2.0/ |
| 6 | CC-BY-NC 2.1 | https://creativecommons.org/licenses/by-nc/2.1/ |
| 7 | CC-BY-NC 2.5 | https://creativecommons.org/licenses/by-nc/2.5/ |
| 8 | CC-BY-NC 3.0 | https://creativecommons.org/licenses/by-nc/3.0/ |
| 9 | CC-BY-NC 4.0 | https://creativecommons.org/licenses/by-nc/4.0/ |
| 10 | CC-BY-NC-ND 2.0 | https://creativecommons.org/licenses/by-nc-nd/2.0/ |
| 11 | CC-BY-NC-ND 2.1 | https://creativecommons.org/licenses/by-nc-nd/2.1/ |
| 12 | CC-BY-NC-ND 2.5 | https://creativecommons.org/licenses/by-nc-nd/2.5/ |
| 13 | CC-BY-NC-ND 3.0 | https://creativecommons.org/licenses/by-nc-nd/3.0/ |
| 14 | CC-BY-NC-ND 4.0 | https://creativecommons.org/licenses/by-nc-nd/4.0/ |
| 15 | CC-BY-NC-SA 2.0 | https://creativecommons.org/licenses/by-nc-sa/2.0/ |
| 16 | CC-BY-NC-SA 2.1 | https://creativecommons.org/licenses/by-nc-sa/2.1/ |
| 17 | CC-BY-NC-SA 2.5 | https://creativecommons.org/licenses/by-nc-sa/2.5/ |
| 18 | CC-BY-NC-SA 3.0 | https://creativecommons.org/licenses/by-nc-sa/3.0/ |
| 19 | CC-BY-NC-SA 4.0 | https://creativecommons.org/licenses/by-nc-sa/4.0/ |
| 20 | CC-BY-ND 2.0 | https://creativecommons.org/licenses/by-nd/2.0/ |
| 21 | CC-BY-ND 2.5 | https://creativecommons.org/licenses/by-nd/2.5/ |
| 22 | CC-BY-ND 3.0 | https://creativecommons.org/licenses/by-nd/3.0/ |
| 23 | CC-BY-ND 4.0 | https://creativecommons.org/licenses/by-nd/4.0/ |
| 24 | CC-BY-SA 2.0 | https://creativecommons.org/licenses/by-sa/2.0/ |
| 25 | CC-BY-SA 2.5 | https://creativecommons.org/licenses/by-sa/2.5/ |
| 26 | CC-BY-SA 3.0 | https://creativecommons.org/licenses/by-sa/3.0/ |
| 27 | CC-BY-SA 4.0 | https://creativecommons.org/licenses/by-sa/4.0/ |
| 28 | CC-NC-Sampling+ 1.0 | https://creativecommons.org/licenses/nc-sampling+/1.0/ |
| 29 | CC-Sampling+ 1.0 | https://creativecommons.org/licenses/sampling+/1.0/ |
| 30 | FMA Sound Recording Common Law | https://freemusicarchive.org/Sound_Recording_Common_Law |
| 31 | Free Art License | https://artlibre.org/licence/lal/en |
| 32 | Free Music Philosophy (FMP) | https://irdial.com/free_and_easy.htm |

## Total Duration by License

| License | Total Duration (Percentage) |
| ------- | --------------------------- |
| CC-BY-NC-SA 3.0 | 291.6 hours (33.55%) |
| CC-BY-NC-ND 3.0 | 237.3 hours (27.30%) |
| CC-BY-NC-ND 4.0 | 100.6 hours (11.57%) |
| CC-BY-NC-SA 4.0 | 57.0 hours (6.56%) |
| CC-BY 4.0 | 41.3 hours (4.75%) |
| CC-BY-NC 3.0 | 38.8 hours (4.47%) |
| CC-BY-NC 4.0 | 28.1 hours (3.23%) |
| CC-BY 3.0 | 15.3 hours (1.76%) |
| CC-BY-SA 4.0 | 12.9 hours (1.48%) |
| CC-BY-SA 3.0 | 9.8 hours(1.13%) |
| CC-BY-NC-SA 2.0 | 7.0 hours (0.81%) |
| CC-BY-NC-ND 2.0 | 6.4 hours (0.74%) |
| CC-BY-ND 3.0 | 4.5 hours (0.52%) |
| FMA Sound Recording Common Law | 3.5 hours (0.40%) |
| CC-BY-ND 4.0 | 3.3 hours (0.38%) |
| CC-BY-NC-ND 2.5 | 2.9 hours (0.33%) |
| CC-BY-NC-SA 2.5 | 2.2 hours (0.25%) |
| CC0 1.0 | 1.3 hours (0.15%) |
| CC-BY-NC 2.5 | 1.2 hours (0.14%) |
| Free Music Philosophy (FMP) | 1.1 hours (0.13%) |
| CC-BY 1.0 | 52.0 minutes (0.10%) |
| CC-BY-SA 2.0 | 24.0 minutes (0.05%) |
| CC-BY 2.0 | 16.5 minutes (0.03%) |
| CC-BY-NC-SA 2.1 | 15.0 minutes (0.03%) |
| CC-BY-NC 2.1 | 15.0 minutes (0.03%) |
| CC-BY-NC 2.0 | 12.0 minutes (0.02%) |
| CC-NC-Sampling+ 1.0 | 9.5 minutes (0.02%) |
| CC-BY-NC-ND 2.1 | 8.5 minutes (0.02%) |
| Free Art License | 7.8 minutes (0.02%) |
| CC-Sampling+ 1.0 | 7.0 minutes (0.01%) |
| CC-BY-ND 2.5 | 5.7 minutes (0.01%) |
| CC-BY-SA 2.5 | 4.5 minutes (0.01%) |
| CC-BY-ND 2.0 | 3.5 minutes (0.01%) |
| CC-BY 2.5 | 1.0 minutes (0.00%) |

# Citations

```
@inproceedings{fma_dataset,
title = {{FMA}: A Dataset for Music Analysis},
author = {Defferrard, Micha\"el and Benzi, Kirell and Vandergheynst, Pierre and Bresson, Xavier},
booktitle = {18th International Society for Music Information Retrieval Conference (ISMIR)},
year = {2017},
archiveprefix = {arXiv},
eprint = {1612.01840},
url = {https://arxiv.org/abs/1612.01840},
}
```

```
@inproceedings{fma_challenge,
title = {Learning to Recognize Musical Genre from Audio},
subtitle = {Challenge Overview},
author = {Defferrard, Micha\"el and Mohanty, Sharada P. and Carroll, Sean F. and Salath\'e, Marcel},
booktitle = {The 2018 Web Conference Companion},
year = {2018},
publisher = {ACM Press},
isbn = {9781450356404},
doi = {10.1145/3184558.3192310},
archiveprefix = {arXiv},
eprint = {1803.05337},
url = {https://arxiv.org/abs/1803.05337},
}
```

# FMA：音乐分析数据集

[迈克尔·德费拉德](https://deff.ch/)、[基雷尔·本齐](https://kirellbenzi.com/)、[皮埃尔·凡德根斯特](https://people.epfl.ch/pierre.vandergheynst)、[泽维尔·布雷松](https://www.ntu.edu.sg/home/xbresson)

**国际音乐信息检索学会会议（ISMIR, 2017）**

> 本文介绍了免费音乐档案馆（Free Music Archive, FMA）——一个开放易用的数据集，适用于评估音乐信息检索（Music Information Retrieval, MIR）领域的多项任务，该领域专注于浏览、检索与组织大规模音乐馆藏。当前学界对特征学习与端到端学习的兴趣日益增长，但受限于大规模音频数据集的匮乏。FMA旨在解决这一难题，共提供来自16341位艺术家、14854张专辑的106574首曲目，总时长343天，容量达917 GiB，所有音频均采用知识共享（Creative Commons）许可协议发布，并按照包含161个类别的分层分类体系进行组织。该数据集不仅提供完整高品质音频与预计算特征，还包含曲目级、用户级元数据、标签以及传记等自由格式文本。本文将详细阐述该数据集的构成与构建流程，提出训练集/验证集/测试集划分方案与三个子集，讨论适用于MIR的部分任务，并针对音乐流派识别任务评估了若干基线模型。代码、数据集与使用示例可访问：https://github.com/mdeff/fma。

论文：[arXiv:1612.01840](https://arxiv.org/abs/1612.01840) - [LaTeX源码与审稿记录](https://github.com/mdeff/paper-fma-ismir2017)

幻灯片：[doi:10.5281/zenodo.1066119](https://doi.org/10.5281/zenodo.1066119)

海报：[doi:10.5281/zenodo.1035847](https://doi.org/10.5281/zenodo.1035847)

# 本数据集包
本为**大型**数据集包，总计包含**105024个样本**，所有音频均被裁剪为**30秒**片段，涵盖**16个**非平衡音乐流派，总音频时长达**869.2小时**。

## 重新打包说明
- 173个文件无法被`libsndfile / libmpg123`读取，已移除。
- 1377个文件的许可协议不明确是否允许重新分发，或完整许可文本不可用，已移除。

## 许可协议
- [FMA代码库](https://github.com/mdeff/fma)采用[MIT许可协议](https://github.com/mdeff/fma/blob/master/LICENSE.txt)发布。
- FMA元数据采用[CC-BY 4.0](https://creativecommons.org/licenses/by/4.0/)许可协议发布。
- 单个音频文件采用多种知识共享系列许可协议，附带少量其他许可协议。**每个文件均附带其许可协议，并枚举了许可协议的关键细节。** 为方便开发者与训练者使用，我们提供了配置选项，可仅限定使用可商用的数据。

如需了解更多细节，请访问以下任一链接。

| 类别标签 | 许可协议名称 | 链接 |
| ----------- | ------------ | --- |
| 0 | CC-BY 1.0 | https://creativecommons.org/licenses/by/1.0/ |
| 1 | CC-BY 2.0 | https://creativecommons.org/licenses/by/2.0/ |
| 2 | CC-BY 2.5 | https://creativecommons.org/licenses/by/2.5/ |
| 3 | CC-BY 3.0 | https://creativecommons.org/licenses/by/3.0/ |
| 4 | CC-BY 4.0 | https://creativecommons.org/licenses/by/4.0/ |
| 5 | CC-BY-NC 2.0 | https://creativecommons.org/licenses/by-nc/2.0/ |
| 6 | CC-BY-NC 2.1 | https://creativecommons.org/licenses/by-nc/2.1/ |
| 7 | CC-BY-NC 2.5 | https://creativecommons.org/licenses/by-nc/2.5/ |
| 8 | CC-BY-NC 3.0 | https://creativecommons.org/licenses/by-nc/3.0/ |
| 9 | CC-BY-NC 4.0 | https://creativecommons.org/licenses/by-nc/4.0/ |
| 10 | CC-BY-NC-ND 2.0 | https://creativecommons.org/licenses/by-nc-nd/2.0/ |
| 11 | CC-BY-NC-ND 2.1 | https://creativecommons.org/licenses/by-nc-nd/2.1/ |
| 12 | CC-BY-NC-ND 2.5 | https://creativecommons.org/licenses/by-nc-nd/2.5/ |
| 13 | CC-BY-NC-ND 3.0 | https://creativecommons.org/licenses/by-nc-nd/3.0/ |
| 14 | CC-BY-NC-ND 4.0 | https://creativecommons.org/licenses/by-nc-nd/4.0/ |
| 15 | CC-BY-NC-SA 2.0 | https://creativecommons.org/licenses/by-nc-sa/2.0/ |
| 16 | CC-BY-NC-SA 2.1 | https://creativecommons.org/licenses/by-nc-sa/2.1/ |
| 17 | CC-BY-NC-SA 2.5 | https://creativecommons.org/licenses/by-nc-sa/2.5/ |
| 18 | CC-BY-NC-SA 3.0 | https://creativecommons.org/licenses/by-nc-sa/3.0/ |
| 19 | CC-BY-NC-SA 4.0 | https://creativecommons.org/licenses/by-nc-sa/4.0/ |
| 20 | CC-BY-ND 2.0 | https://creativecommons.org/licenses/by-nd/2.0/ |
| 21 | CC-BY-ND 2.5 | https://creativecommons.org/licenses/by-nd/2.5/ |
| 22 | CC-BY-ND 3.0 | https://creativecommons.org/licenses/by-nd/3.0/ |
| 23 | CC-BY-ND 4.0 | https://creativecommons.org/licenses/by-nd/4.0/ |
| 24 | CC-BY-SA 2.0 | https://creativecommons.org/licenses/by-sa/2.0/ |
| 25 | CC-BY-SA 2.5 | https://creativecommons.org/licenses/by-sa/2.5/ |
| 26 | CC-BY-SA 3.0 | https://creativecommons.org/licenses/by-sa/3.0/ |
| 27 | CC-BY-SA 4.0 | https://creativecommons.org/licenses/by-sa/4.0/ |
| 28 | CC-NC-Sampling+ 1.0 | https://creativecommons.org/licenses/nc-sampling+/1.0/ |
| 29 | CC-Sampling+ 1.0 | https://creativecommons.org/licenses/sampling+/1.0/ |
| 30 | FMA Sound Recording Common Law | https://freemusicarchive.org/Sound_Recording_Common_Law |
| 31 | Free Art License | https://artlibre.org/licence/lal/en |
| 32 | Free Music Philosophy (FMP) | https://irdial.com/free_and_easy.htm |

## 各许可协议总时长

| 许可协议 | 总时长（占比） |
| ------- | --------------------------- |
| CC-BY-NC-SA 3.0 | 291.6小时（33.55%） |
| CC-BY-NC-ND 3.0 | 237.3小时（27.30%） |
| CC-BY-NC-ND 4.0 | 100.6小时（11.57%） |
| CC-BY-NC-SA 4.0 | 57.0小时（6.56%） |
| CC-BY 4.0 | 41.3小时（4.75%） |
| CC-BY-NC 3.0 | 38.8小时（4.47%） |
| CC-BY-NC 4.0 | 28.1小时（3.23%） |
| CC-BY 3.0 | 15.3小时（1.76%） |
| CC-BY-SA 4.0 | 12.9小时（1.48%） |
| CC-BY-SA 3.0 | 9.8小时（1.13%） |
| CC-BY-NC-SA 2.0 | 7.0小时（0.81%） |
| CC-BY-NC-ND 2.0 | 6.4小时（0.74%） |
| CC-BY-ND 3.0 | 4.5小时（0.52%） |
| FMA Sound Recording Common Law | 3.5小时（0.40%） |
| CC-BY-ND 4.0 | 3.3小时（0.38%） |
| CC-BY-NC-ND 2.5 | 2.9小时（0.33%） |
| CC-BY-NC-SA 2.5 | 2.2小时（0.25%） |
| CC0 1.0 | 1.3小时（0.15%） |
| CC-BY-NC 2.5 | 1.2小时（0.14%） |
| Free Music Philosophy (FMP) | 1.1小时（0.13%） |
| CC-BY 1.0 | 52.0分钟（0.10%） |
| CC-BY-SA 2.0 | 24.0分钟（0.05%） |
| CC-BY 2.0 | 16.5分钟（0.03%） |
| CC-BY-NC-SA 2.1 | 15.0分钟（0.03%） |
| CC-BY-NC 2.1 | 15.0分钟（0.03%） |
| CC-BY-NC 2.0 | 12.0分钟（0.02%） |
| CC-NC-Sampling+ 1.0 | 9.5分钟（0.02%） |
| CC-BY-NC-ND 2.1 | 8.5分钟（0.02%） |
| Free Art License | 7.8分钟（0.02%） |
| CC-Sampling+ 1.0 | 7.0分钟（0.01%） |
| CC-BY-ND 2.5 | 5.7分钟（0.01%） |
| CC-BY-SA 2.5 | 4.5分钟（0.01%） |
| CC-BY-ND 2.0 | 3.5分钟（0.01%） |
| CC-BY 2.5 | 1.0分钟（0.00%） |

## 引用格式

@inproceedings{fma_dataset,
title = {{FMA}: A Dataset for Music Analysis},
author = {Defferrard, Michaël and Benzi, Kirell and Vandergheynst, Pierre and Bresson, Xavier},
booktitle = {18th International Society for Music Information Retrieval Conference (ISMIR)},
year = {2017},
archiveprefix = {arXiv},
eprint = {1612.01840},
url = {https://arxiv.org/abs/1612.01840},
}

@inproceedings{fma_challenge,
title = {Learning to Recognize Musical Genre from Audio},
subtitle = {Challenge Overview},
author = {Defferrard, Michaël and Mohanty, Sharada P. and Carroll, Sean F. and Salathé, Marcel},
booktitle = {The 2018 Web Conference Companion},
year = {2018},
publisher = {ACM Press},
isbn = {9781450356404},
doi = {10.1145/3184558.3192310},
archiveprefix = {arXiv},
eprint = {1803.05337},
url = {https://arxiv.org/abs/1803.05337},
}

搜集汇总

背景与挑战

背景概述

该数据集是FMA音乐分析数据集的一个大规模子集，包含105,024个30秒音频样本，覆盖16个不平衡流派，总时长约869.2小时。所有音频采用Creative Commons系列许可，适用于音乐信息检索任务。