24,800 AI-generated 20-second music clips created using the facebook/musicgen-small model. The dataset is the audio companion to Prompt2MusicBench, with each clip generated from a structured text prompt encoding genre, instrument, tempo, and mood. It was created by bodhisattamaiti and last updated on Hugging Face in December 2025.
Use Cases
- Training or evaluating text-to-music models based on structured prompt inputs.
- Analyzing the relationship between textual attributes (genre, mood) and generated audio features.
- Benchmarking audio generation quality across different prompt conditions.
- Studying the stylistic output of the facebook/musicgen-small model.
Strengths
- Contains 24,800 audio samples, providing a substantial collection for analysis.
- Each 20-second clip is generated from a structured prompt with defined attributes like genre and tempo.
- Linked to a companion benchmark dataset (Prompt2MusicBench) for structured evaluation.
Limitations
- Column-level documentation is absent; field semantics must be inferred after download.
- Row count is known, but file formats, size, and license information are unknown.
- Data may reflect the stylistic and technical biases inherent to the facebook/musicgen-small model used for generation.
Provenance
- Source
- Hugging Face, created by bodhisattamaiti.
- Collection Method
- AI-generated using the facebook/musicgen-small model from structured text prompts.
- Time Range
- null
- Freshness
- Last updated 2025-12-10 13:14:14; freshness should be verified.
- Geography
- null