KsponSpeech¶

class openspeech.datasets.ksponspeech.lit_data_module.LightningKsponSpeechDataModule(*args: Any, **kwargs: Any)[source]¶

Lightning data module for KsponSpeech. KsponSpeech corpus contains 969 h of general open-domain dialog utterances, spoken by about 2000 native Korean speakers in a clean environment. All data were constructed by recording the dialogue of two people freely conversing on a variety of topics and manually transcribing the utterances. The transcription provides a dual transcription consisting of orthography and pronunciation, and disfluency tags for spontaneity of speech, such as filler words, repeated words, and word fragments.

KSPONSPEECH_TRAIN_NUM¶

the number of KsponSpeech’s train data.

Type: int

Parameters: configs (DictConfig) – configuration set.

prepare_data()[source]¶

Prepare KsponSpeech manifest file. If there is not exist manifest file, generate manifest file.

Returns: tokenizer is in charge of preparing the inputs for a model.
Return type: tokenizer (Tokenizer)

setup(stage: Optional[str] = None, tokenizer: openspeech.tokenizers.tokenizer.Tokenizer = None)[source]¶

Split train and valid dataset for training.

Parameters

stage (str) – stage of training. train or valid
tokenizer (Tokenizer) – tokenizer is in charge of preparing the inputs for a model.

Returns

None

test_dataloader() → openspeech.data.audio.data_loader.AudioDataLoader [source]¶: Return data loader for training.

train_dataloader() → openspeech.data.audio.data_loader.AudioDataLoader [source]¶: Return data loader for training.

val_dataloader() → openspeech.data.audio.data_loader.AudioDataLoader [source]¶: Return data loader for validation.