espnet2.asr.espnet_model.ESPnetASRModel

source

class espnet2.asr.espnet_model.ESPnetASRModel(vocab_size: int, token_list: Tuple[str, ...] | List[str], frontend: AbsFrontend | None, specaug: AbsSpecAug | None, normalize: AbsNormalize | None, preencoder: AbsPreEncoder | None, encoder: AbsEncoder, postencoder: AbsPostEncoder | None, decoder: AbsDecoder | None, ctc: CTC, joint_network: Module | None, aux_ctc: dict | None = None, ctc_weight: float = 0.5, interctc_weight: float = 0.0, ignore_id: int = -1, lsm_weight: float = 0.0, length_normalized_loss: bool = False, report_cer: bool = True, report_wer: bool = True, sym_space: str = '<space>', sym_blank: str = '<blank>', transducer_multi_blank_durations: List = [], transducer_multi_blank_sigma: float = 0.05, sym_sos: str = '<sos/eos>', sym_eos: str = '<sos/eos>', extract_feats_in_collect_stats: bool = True, lang_token_id: int = -1)

Bases: AbsESPnetModel

CTC-attention hybrid Encoder-Decoder model

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

batchify_nll(encoder_out: Tensor, encoder_out_lens: Tensor, ys_pad: Tensor, ys_pad_lens: Tensor, batch_size: int = 100)

Compute negative log likelihood(nll) from transformer-decoder

To avoid OOM, this fuction seperate the input into batches. Then call nll for each batch and combine and return results. :param encoder_out: (Batch, Length, Dim) :param encoder_out_lens: (Batch,) :param ys_pad: (Batch, Length) :param ys_pad_lens: (Batch,) :param batch_size: int, samples each batch contain when computing nll,

you may change this to avoid OOM or increase GPU memory usage

collect_feats(speech: Tensor, speech_lengths: Tensor, text: Tensor, text_lengths: Tensor, **kwargs) → Dict[str, Tensor]

encode(speech: Tensor, speech_lengths: Tensor) → Tuple[Tensor, Tensor]

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

• Parameters:
  • speech – (Batch, Length, …)
  • speech_lengths – (Batch, )

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

Frontend + Encoder + Decoder + Calc loss

• Parameters:
  • speech – (Batch, Length, …)
  • speech_lengths – (Batch, )
  • text – (Batch, Length)
  • text_lengths – (Batch,)
  • kwargs – “utt_id” is among the input.

nll(encoder_out: Tensor, encoder_out_lens: Tensor, ys_pad: Tensor, ys_pad_lens: Tensor) → Tensor

Compute negative log likelihood(nll) from transformer-decoder

Normally, this function is called in batchify_nll.

• Parameters:
  • encoder_out – (Batch, Length, Dim)
  • encoder_out_lens – (Batch,)
  • ys_pad – (Batch, Length)
  • ys_pad_lens – (Batch,)