The FIU Research Data Portal supports finding, sharing, and publishing finalized university research datasets, including code, documentation, and metadata.
Dataset
Million Song Dataset
The Million Song Dataset provides audio features and metadata for popular music tracks to support research and evaluation.
About Million Song Dataset
The Million Song Dataset is a freely available collection of audio features and metadata for one million contemporary popular music tracks, with core analysis provided by The Echo Nest. It also gathers complementary datasets covering cover songs, lyrics, tags and similarity, user data, and genre labels.
The dataset does not include audio, only derived features, though sample audio can be fetched from services like 7digital using provided code. The full download is 280 GB, with a 10,000-song subset available as a 1.8 GB compressed download.
Key features
- Audio features and metadata for one million tracks
- SecondHandSongs data identifying cover songs
- musiXmatch dataset with lyrics
- Last.fm data for tags and song similarity
- User data from Taste Profile subset and thisismyjam mapping
- Genre labels from tagtraum and Top MAGD datasets
Best for: Researchers in music information retrieval who need large-scale features and metadata for evaluation.
Good for
- Freely available collection for research and evaluation
- Combines core features with community-contributed lyric, tag, user, and genre datasets
- Provides a smaller 10,000-song subset and example record for initial exploration
Watch out for
- Contains no audio, only derived features and metadata
- Full download requires 280 GB
- Forum participation requires registration
Million Song Dataset at a glance
- Official website
- millionsongdataset.com
- Category
- Research & Data › Open Data & Datasets
- Type
- Dataset
- Platforms
- Web
- Language
- English
- Last checked
- Oct 5, 2026 · Online when checked
- Listed
- Oct 5, 2026
Similar data sources
Other listings in Open Data & Datasets with the most in common.
USAP-DC supports Antarctic project and dataset search, metadata submission, and access to data management resources.
Dataset
The THREDDS Data Server user guide documents setup, configuration, catalogs, metadata, and related server services.
Dataset
The Image Data Resource makes published research datasets available for searching, viewing, analysis, and processing.
Dataset
CROSSDA is a Croatian social science data archive with research dataverses and dataset download records.
Dataset
Fairdata Services offers tools to store, find, access, describe, and share research data, including the IDA storage service and Etsin dataset finder.
Dataset
Popular in Open Data & Datasets
Well-known listings from the same topic.
data.worlddata.worldNASA APIapi.nasa.govCosmoSimcosmosim.orgdata.meudon.frdata.meudon.frESS-DIVEess-dive.lbl.govState of Hawaiidata.hawaii.gov
Million Song Dataset FAQ
Does the Million Song Dataset include audio?
No. It contains only derived features and metadata, with code provided to fetch sample audio from services like 7digital.
How large are the downloads?
The full dataset is 280 GB, and a 10,000-song subset is available as a 1.8 GB compressed download.
How should I cite the dataset?
Cite Thierry Bertin-Mahieux, Daniel P.W. Ellis, Brian Whitman, and Paul Lamere, The Million Song Dataset, Proceedings of ISMIR 2011.
Is something wrong or out of date?
Tell us and an editor will check it.
Own this website?
Claim this listing
Claim the listing and suggest updates yourself. An editor checks every claim first.