MusicCaps is a dataset composed of 5.5k music-text pairs, with rich text descriptions provided by human experts. For each 10-second music clip, MusicCaps provides:
37 PAPERS • 1 BENCHMARK
The Song Describer Dataset (SDD) contains ~1.1k captions for 706 permissively licensed music recordings. It is designed for use in evaluation of models that address music-and-language (M&L) tasks such as music captioning, text-to-music generation and music-language retrieval.
1 PAPER • NO BENCHMARKS YET