| | |
|---|
| | | self.predictor_bias = predictor_bias |
|---|
| | | self.sampling_ratio = sampling_ratio |
|---|
| | | self.criterion_pre = mae_loss(normalize_length=length_normalized_loss) |
|---|
| | | # self.step_cur = 0 |
|---|
| | | # |
|---|
| | | |
|---|
| | | |
|---|
| | | self.share_embedding = share_embedding |
|---|
| | | if self.share_embedding: |
|---|
| | | self.decoder.embed = None |
|---|
| | |
|---|
| | | speech, speech_lengths = data_in, data_lengths |
|---|
| | | if len(speech.shape) < 3: |
|---|
| | | speech = speech[None, :, :] |
|---|
| | | if speech_lengths is None: |
|---|
| | | if speech_lengths is not None: |
|---|
| | | speech_lengths = speech_lengths.squeeze(-1) |
|---|
| | | else: |
|---|
| | | speech_lengths = speech.shape[1] |
|---|
| | | else: |
|---|
| | | # extract fbank feats |
|---|
| | |
|---|
| | | |
|---|
| | | result_i = {"key": key[i], "text": text_postprocessed} |
|---|
| | | |
|---|
| | | |
|---|
| | | if ibest_writer is not None: |
|---|
| | | ibest_writer["token"][key[i]] = " ".join(token) |
|---|
| | | # ibest_writer["text"][key[i]] = text |
|---|
