espnet2.speechlm.dataloader.dataset.SingleDataset

source

class espnet2.speechlm.dataloader.dataset.SingleDataset(json_file: str, rank: int = 0, world_size: int = 1)

Bases: Dataset

ESPnet Speech Language Model Dataset.

• Parameters:
  • json_file – Path to dataset JSON created by prepare_dataset_json.py
  • rank – Process rank for distributed training (default: 0)
  • world_size – Total number of processes (default: 1)

property entries : List[str]

Return list of all data entry names.

property sample_ids : List[str]

Return list of all sample IDs.