espnet3.systems.asr.task.ASRTask

source

class espnet3.systems.asr.task.ASRTask

Bases: AbsTask

ASR task definition for argument wiring and model construction.

classmethod add_task_arguments(parser: ArgumentParser)

Add ASR-specific task arguments to the parser.

classmethod build_collate_fn(args: Namespace, train: bool) → Callable[[Collection[Tuple[str, Dict[str, ndarray]]]], Tuple[List[str], Dict[str, Tensor]]]

Build the collate function for ASR batches.

classmethod build_model(args: Namespace) → ESPnetASRModel

Build the ESPnet ASR model from parsed arguments.

classmethod build_preprocess_fn(args: Namespace, train: bool) → Callable[[str, Dict[str, array]], Dict[str, ndarray]] | None

Build the preprocessing function for ASR inputs.

class_choices_list : List[[ClassChoices](../../espnet2/train/ClassChoices.md#espnet2.train.class_choices.ClassChoices)] = [<espnet2.train.class_choices.ClassChoices object>, <espnet2.train.class_choices.ClassChoices object>, <espnet2.train.class_choices.ClassChoices object>, <espnet2.train.class_choices.ClassChoices object>, <espnet2.train.class_choices.ClassChoices object>, <espnet2.train.class_choices.ClassChoices object>, <espnet2.train.class_choices.ClassChoices object>, <espnet2.train.class_choices.ClassChoices object>, <espnet2.train.class_choices.ClassChoices object>]

num_optimizers : int = 1

classmethod optional_data_names(train: bool = True, inference: bool = False) → Tuple[str, ...]

Return optional data field names for the task.

classmethod required_data_names(train: bool = True, inference: bool = False) → Tuple[str, ...]

Return required data field names for the task.

trainer

alias of Trainer