LP-MusicCaps: Large Language Model Generated Music Captions

LP-MusicCaps-MTT is a dataset of pseudo music captions generated by a Large Language Model for text-to-music and music-to-text tasks. The dataset was constructed by combining three existing multi-label tag datasets and four task-specific datasets. It was created by seungheondoh and last updated on August 4, 2023.

Use Cases

• Train text-to-music generation models based on the LLM-generated caption pairs.
• Train music-to-text retrieval models based on the music-to-caption pairs.
• Benchmark music captioning models using the pseudo captions.
• Analyze the relationship between music tags and descriptive captions.
• Develop multimodal audio-text models leveraging the paired data.

Strengths

• Dataset leverages three existing multi-label tag datasets, suggesting a foundation in established music annotation.
• Dataset combines four task-specific datasets, indicating a multi-purpose construction.
• Dataset was last updated on August 4, 2023, providing a specific timestamp.

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.
• Row count is unknown, which may limit suitability assessment.

Provenance

Source

huggingface

Collection Method

Constructed by combining three existing multi-label tag datasets and four task-specific datasets.

Freshness

Last updated 2023-08-04 10:39:28; freshness should be verified.