| | |
|---|
| | | import numpy as np |
|---|
| | | |
|---|
| | | import torch |
|---|
| | | from typeguard import check_argument_types |
|---|
| | | |
|---|
| | | from funasr.layers.abs_normalize import AbsNormalize |
|---|
| | | from funasr.models.ctc import CTC |
|---|
| | |
|---|
| | | preencoder: Optional[AbsPreEncoder] = None, |
|---|
| | | postencoder: Optional[AbsPostEncoder] = None, |
|---|
| | | ): |
|---|
| | | assert check_argument_types() |
|---|
| | | assert 0.0 <= ctc_weight <= 1.0, ctc_weight |
|---|
| | | assert 0.0 <= interctc_weight < 1.0, interctc_weight |
|---|
| | | |
|---|
| | |
|---|
| | | else: |
|---|
| | | hw_embed = self.bias_embed(hw_list_pad) |
|---|
| | | hw_embed, (h_n, _) = self.bias_encoder(hw_embed) |
|---|
| | | hw_embed = h_n.repeat(encoder_out.shape[0], 1, 1) |
|---|
| | | else: |
|---|
| | | hw_lengths = [len(i) for i in hw_list] |
|---|
| | | hw_list_pad = pad_list([torch.Tensor(i).long() for i in hw_list], 0).to(encoder_out.device) |
|---|
