Free Music Archive (FMA)
收藏资源简介:
Free Music Archive (FMA) 是一个由瑞士洛桑联邦理工学院和新加坡南洋理工大学创建的大型音乐分析数据集,包含106,574条来自16,341位艺术家的音乐曲目,总数据量达917 GiB,涵盖161种音乐类型。数据集通过Creative Commons许可证授权,提供完整的音频文件和高质量的预计算特征,以及详细的曲目和用户级元数据。FMA旨在支持音乐信息检索(MIR)领域的多种任务,如音乐分类、自动标注和音乐结构分析等,特别适用于深度学习模型的训练和评估。
Free Music Archive (FMA) is a large-scale music analysis dataset developed by École Polytechnique Fédérale de Lausanne in Switzerland and Nanyang Technological University in Singapore. It contains 106,574 music tracks from 16,341 artists, with a total data volume of 917 GiB and coverage of 161 music genres. Licensed under Creative Commons licenses, the dataset provides complete audio files, high-quality pre-computed features, as well as detailed track-level and user-level metadata. FMA is designed to support various tasks in the field of Music Information Retrieval (MIR), such as music classification, automatic annotation, music structure analysis and etc., and is particularly suitable for the training and evaluation of deep learning models.




