机器学习与语料Free to access
Key facts
Contents & fields
The dataset contains metadata for over 200,000 songs, each record including song ID, title, artist, release year, album, genre, tags, features, views, duration, language detection results, and artist biographical data from Wikidata (birth year, birth country, sex, band status, awards). Data is provided in CSV format.
- id——Unique song identifier
- title——Song title
- artist——Artist or band name
- year——Release year
- album——Album containing the song
- genre_song——Primary music genre
- tag——Associated tags or keywords
- features——Musical or technical features
- views——Number of views or plays
- duration_s——Song duration in seconds
- language_cld3——Language detected by CLD3
- language_ft——Language detected by FastText
Research uses
Suitable for music information retrieval, music recommendation systems, artist profiling, lyric language identification, and music trend analysis.
This card was drafted from the source page; institution, coverage, time span, scale, fields and license are subject to the official page (pending human review).
Keywords
Why this is hard to get on your own
Obtaining large-scale structured music metadata with artist background, but note the non-commercial license restriction.