update

游雁

2024-01-05 622d799de8032f957c0d335827fdc637b8d89078

 funasr/models/monotonic_aligner/model.py

@@ -4,11 +4,13 @@
from torch.cuda.amp import autocast
from typing import Union, Dict, List, Tuple, Optional

from funasr.register import tables
from funasr.utils import postprocess_utils
from funasr.utils.datadir_writer import DatadirWriter
from funasr.models.paraformer.cif_predictor import mae_loss
from funasr.models.transformer.utils.add_sos_eos import add_sos_eos
from funasr.models.transformer.utils.nets_utils import make_pad_mask, pad_list
from funasr.metrics.compute_acc import th_accuracy
from funasr.train_utils.device_funcs import force_gatherable
from funasr.models.transformer.utils.add_sos_eos import add_sos_eos
from funasr.models.transformer.utils.nets_utils import make_pad_mask
from funasr.utils.timestamp_tools import ts_prediction_lfr6_standard
from funasr.utils import postprocess_utils
from funasr.utils.datadir_writer import DatadirWriter
@@ -25,7 +27,6 @@
    Achieving timestamp prediction while recognizing with non-autoregressive end-to-end ASR model
    https://arxiv.org/abs/2301.12343
    """

    def __init__(
        self,
        input_size: int = 80,
@@ -41,7 +42,6 @@
        length_normalized_loss: bool = False,
        **kwargs,
    ):

        super().__init__()

        if specaug is not None:
@@ -155,7 +155,6 @@
             frontend=None,
             **kwargs,
             ):
        
        meta_data = {}
        # extract fbank feats
        time1 = time.perf_counter()
@@ -190,8 +189,7 @@
            timestamp_str, timestamp = ts_prediction_lfr6_standard(us_alpha[:encoder_out_lens[i] * 3],
                                                                   us_peak[:encoder_out_lens[i] * 3],
                                                                   copy.copy(token))
            text_postprocessed, time_stamp_postprocessed, word_lists = postprocess_utils.sentence_postprocess(
                token, timestamp)
            text_postprocessed, time_stamp_postprocessed, _ = postprocess_utils.sentence_postprocess(token, timestamp)
            result_i = {"key": key[i], "text": text_postprocessed,
                                "timestamp": time_stamp_postprocessed,
                                }