| | |
|---|
| | | results = speech2text(**batch) |
|---|
| | | if len(results) < 1: |
|---|
| | | hyp = Hypothesis(score=0.0, scores={}, states={}, yseq=[]) |
|---|
| | | results = [[" ", ["sil"], [2], hyp, 10, 6]] * nbest |
|---|
| | | results = [[" ", ["sil"], [2], hyp, 10, 6, []]] * nbest |
|---|
| | | time_end = time.time() |
|---|
| | | forward_time = time_end - time_beg |
|---|
| | | lfr_factor = results[0][-1] |
|---|
| | |
|---|
| | | logging.info(rtf_avg) |
|---|
| | | if writer is not None: |
|---|
| | | ibest_writer["rtf"]["rtf_avf"] = rtf_avg |
|---|
| | | torch.cuda.empty_cache() |
|---|
| | | return asr_result_list |
|---|
| | | |
|---|
| | | return _forward |
|---|
| | |
|---|
| | | ibest_writer["time_stamp"][key] = "{}".format(time_stamp_postprocessed) |
|---|
| | | |
|---|
| | | logging.info("decoding, utt: {}, predictions: {}".format(key, text_postprocessed_punc)) |
|---|
| | | torch.cuda.empty_cache() |
|---|
| | | return asr_result_list |
|---|
| | | |
|---|
| | | return _forward |
|---|
| | |
|---|
| | | format="%(asctime)s (%(module)s:%(lineno)d) %(levelname)s: %(message)s", |
|---|
| | | ) |
|---|
| | | |
|---|
| | | if ngpu >= 1: |
|---|
| | | if ngpu >= 1 and torch.cuda.is_available(): |
|---|
| | | device = "cuda" |
|---|
| | | else: |
|---|
| | | device = "cpu" |
|---|
| | |
|---|
| | | left_context=left_context, |
|---|
| | | right_context=right_context, |
|---|
| | | ) |
|---|
| | | speech2text = Speech2TextTransducer.from_pretrained( |
|---|
| | | model_tag=model_tag, |
|---|
| | | **speech2text_kwargs, |
|---|
| | | ) |
|---|
| | | speech2text = Speech2TextTransducer(**speech2text_kwargs) |
|---|
| | | |
|---|
| | | def _forward(data_path_and_name_and_type, |
|---|
| | | raw_inputs: Union[np.ndarray, torch.Tensor] = None, |
|---|
