espnet2.asr.discrete_asr_espnet_model.ESPnetDiscreteASRModel

source

class espnet2.asr.discrete_asr_espnet_model.ESPnetDiscreteASRModel(vocab_size: int, token_list: Tuple[str, ...] | List[str], frontend: AbsFrontend | None, specaug: AbsSpecAug | None, preencoder: AbsPreEncoder | None, encoder: AbsEncoder, postencoder: AbsPostEncoder | None, decoder: AbsDecoder, ctc: CTC | None, ctc_weight: float = 0.5, interctc_weight: float = 0.0, src_vocab_size: int = 0, src_token_list: Tuple[str, ...] | List[str] = [], ignore_id: int = -1, lsm_weight: float = 0.0, length_normalized_loss: bool = False, report_bleu: bool = True, sym_space: str = '<space>', sym_blank: str = '<blank>', patch_size: int = 1, extract_feats_in_collect_stats: bool = True, share_decoder_input_output_embed: bool = False, share_encoder_decoder_input_embed: bool = False)

Bases: ESPnetMTModel

Encoder-Decoder model

Initializes internal Module state, shared by both nn.Module and ScriptModule.

encode(src_text: Tensor, src_text_lengths: Tensor) → Tuple[Tensor, Tensor]

Frontend + Encoder. Note that this method is used by mt_inference.py

• Parameters:
  • src_text – (Batch, Length, …)
  • src_text_lengths – (Batch, )

forward(text: Tensor, text_lengths: Tensor, src_text: Tensor, src_text_lengths: Tensor, **kwargs) → Tuple[Tensor, Dict[str, Tensor], Tensor]

Frontend + Encoder + Decoder + Calc loss

• Parameters:
  • text – (Batch, Length)
  • text_lengths – (Batch,)
  • src_text – (Batch, length)
  • src_text_lengths – (Batch,)
  • kwargs – “utt_id” is among the input.