The [documentation](https://pytorch.org/audio/datasets.html) for torchaudio does not include `bg_iterator` and `disk_cache_iterator` from [code](https://github.com/pytorch/audio/blob/master/torchaudio/datasets/utils.py). Thanks to [comment](https://discuss.pytorch.org/t/audio-dataset-load-large-file-into-memory-in-background/85943) to point this out :)