AI Music Large: 20,000 AI and Human Music Audio Files

A dataset of 20,000 audio files, split evenly between AI-generated and human-composed music. The AI-generated portion comprises 256 files from SunoCaps, 4,872 from Udio, and 4,872 from MusicSet. It was created by SleepyJesse and last updated on December 11, 2024.

Use Cases

• Train AI music detection classifiers based on the labeled AI vs. human source.
• Analyze acoustic features to differentiate between AI-generated and human-composed music.
• Benchmark generative music models using a large, labeled collection of AI outputs.
• Study the stylistic characteristics of music from specific AI sources like Udio and SunoCaps.

Strengths

• Contains 20,000 audio files, providing a substantial corpus for analysis.
• Explicitly labels the source of each file (AI or human) and the specific AI platform for 10,000 files.
• The 'large' variant is approximately 70GiB in size, indicating high-fidelity audio content.

Limitations

• Column-level documentation is absent; field semantics must be inferred after download.
• The distribution of human-composed music sources is not detailed in the provided description.
• Description metadata is limited; actual data quality requires manual inspection after download.

Provenance

Source

Aggregated from AI platforms SunoCaps, Udio, and MusicSet, and an unspecified source of human-composed music.

Collection Method

Likely collected via API or web scraping from the listed AI music generation platforms.

Freshness

Last updated 2024-12-11 23:51:28; freshness should be verified.