datasets
Training and evaluation data, with the modality, task and licence stated up front. Listed live from the Hugging Face Hub.
podcast-tokenized-bg2.5-enj4.5podcast-tokenized-bg3.5-enj5podcast_pile_2The bigger dataset (compressed, uncompressed is a few hundred GB so make sure you have enough storage)
Stats (preliminary):
============================================================
Feed Analysis
============================================================
Total feeds: 4,059,278
Valid feeds: 4,022,099 (99.08%)
Invalid feeds: 37,179 (0.92%)
============================================================
Episode Statistics (Valid Feeds Only)… See the full description on the dataset page: https://huggingface.co/datasets/mrfakename/podcast_pile_2.