FMA: A Dataset For Music Analysis

FMA features 106,574 tracks and includes song title, album, artist, genres; play counts, favorites, comments; description, biography, tags; together with audio (343 days, 917 GiB) and features.



作者 MCI Machine Learning Repository
最後更新 八月 18, 2019, 17:01 (CST)
建立 九月 6, 2018, 16:13 (CST)
Area "Computer"
Associated Tasks "Classification
Attribute Characteristics "Real"
Data Set Characteristics "Multivariate
Date Donated "2017-05-24"
Missing Values "N/A"
Number of Instances "106574"
Number of Web Hits "44478"
Number_of_Attributes "518"


  • YearPredictionMSD

    付費方式 免費
    更新頻率 不定期
    Prediction of the release year of a song from audio features. Songs are mostly western, commercial tracks ranging from 1922 to 2011, with a peak in the year 2000s.
  • LSVT Voice Rehabilitation

    付費方式 免費
    更新頻率 不定期
    126 samples from 14 participants, 309 features. Aim
  • Geographical Original of Music

    付費方式 免費
    更新頻率 不定期
    Instances in this dataset contain audio features extracted from 1059 wave files. The task associated with the data is to predict the geographical origin of music.
  • 林務局所轄自然步道軌跡圖-071_香山森林步道

    付費方式 免費
    更新頻率 不定期
  • 「跨行通匯-交易縣市分佈」結構比統計(月報)

    付費方式 免費
    更新頻率 不定期