Sign in to view source links and access this dataset
Description
1,429,734 AI-generated music tracks sourced from three platforms: Suno, Mureka, and Udio. The dataset is annotated with captions, transcriptions, embeddings, aesthetics scores, and NSFW safety labels. It was created by LAION and last updated on Hugging Face in May 2026.
Use Cases
Training or evaluating music generation models based on the large collection of AI-generated audio.
Developing music search and recommendation engines based on the provided text captions and vector embeddings.
Researching content safety and moderation in AI-generated audio based on the NSFW safety labels.
Analyzing the relationship between audio features and subjective quality based on the aesthetics scores.
Strengths
Large scale with 1,429,734 total music tracks.
Includes annotations from multiple sources, such as captions, transcriptions, and safety labels.
Provides a built-in search engine with both simple and advanced user interfaces.
Limitations
Description metadata is limited; actual data quality requires manual inspection after download.
Column-level documentation is absent; field semantics must be inferred after download.
Data may reflect source bias inherent to the three specific AI music platforms used.
Provenance
Source
LAION, aggregated from Suno, Mureka, and Udio platforms.
Collection Method
Likely collected via API or web scraping from the specified AI music generation platforms.
Freshness
Last updated 2026-05-07 12:03:58; freshness should be verified.
License is unknown, which may restrict commercial or research use.