This dataset is the vocalsound dataset, formatted in the WebDataset format. WebDataset files are essentially tar archives, where each example in the dataset is represented by a pair of files: a WAV audio file and a corresponding JSON metadata file. The JSON file contains the class label and other relevant information for that particular audio sample. $ tar tvf train-000000.tar|head -r--r--r-- bigdata/bigdata 19 2025-01-15 15:45 m2246_0_sneeze.json -r--r--r-- bigdata/bigdata 150230 2025-01-15 15:45 m2246_0_sneeze.wav -r--r--r-- bigdata/bigdata 18 2025-01-15 15:45 m2484_0_sniff.json -r--r--r-- bigdata/bigdata 150230 2025-01-15 15:45 m2484_0_sniff.wav -r--r--r-- bigdata/bigdata 18 2025-01-15 15:45 m0564_0_sniff.json -r--r--r-- bigdata/bigdata 125654 2025-01-15 15:45 m0564_0_sniff.wav -r--r--r-- bigdata/bigdata 18 2025-01-15 15:45 m0967_0_sniff.json -r--r--r-- bigdata/bigdata 213036 2025-01-15 15:45 m0967_0_sniff.wav -r--r--r-- bigdata/bigdata 21 2025-01-15 15:45 f2302_0_laughter.json -r--r--r-- bigdata/bigdata 161154 2025-01-15 15:45 f2302_0_laughter.wav $ cat m2246_0_sneeze.json {"label": "sneeze"}