bug fix

shixian.shi

2024-01-12 c3c78fc5e790d48b3a2f9da79199320c06108d38

 funasr/models/paraformer_streaming/model.py

@@ -19,7 +19,7 @@
import time
# from funasr.layers.abs_normalize import AbsNormalize
from funasr.losses.label_smoothing_loss import (
   LabelSmoothingLoss,  # noqa: H301
    LabelSmoothingLoss,  # noqa: H301
)

from funasr.models.paraformer.cif_predictor import mae_loss
@@ -32,12 +32,12 @@
from funasr.models.paraformer.search import Hypothesis

if LooseVersion(torch.__version__) >= LooseVersion("1.6.0"):
   from torch.cuda.amp import autocast
    from torch.cuda.amp import autocast
else:
   # Nothing to do if torch<1.6.0
   @contextmanager
   def autocast(enabled=True):
      yield
    # Nothing to do if torch<1.6.0
    @contextmanager
    def autocast(enabled=True):
        yield
from funasr.utils.load_utils import load_audio_text_image_video, extract_fbank
from funasr.utils import postprocess_utils
from funasr.utils.datadir_writer import DatadirWriter
@@ -50,531 +50,532 @@

@tables.register("model_classes", "ParaformerStreaming")
class ParaformerStreaming(Paraformer):
   """
   Author: Speech Lab of DAMO Academy, Alibaba Group
   Paraformer: Fast and Accurate Parallel Transformer for Non-autoregressive End-to-End Speech Recognition
   https://arxiv.org/abs/2206.08317
   """
	
   def __init__(
      self,
      *args,
      **kwargs,
   ):
		
      super().__init__(*args, **kwargs)
		
      # import pdb;
      # pdb.set_trace()
      self.sampling_ratio = kwargs.get("sampling_ratio", 0.2)
    """
    Author: Speech Lab of DAMO Academy, Alibaba Group
    Paraformer: Fast and Accurate Parallel Transformer for Non-autoregressive End-to-End Speech Recognition
    https://arxiv.org/abs/2206.08317
    """
    
    def __init__(
        self,
        *args,
        **kwargs,
    ):
        
        super().__init__(*args, **kwargs)
        
        # import pdb;
        # pdb.set_trace()
        self.sampling_ratio = kwargs.get("sampling_ratio", 0.2)


      self.scama_mask = None
      if hasattr(self.encoder, "overlap_chunk_cls") and self.encoder.overlap_chunk_cls is not None:
         from funasr.models.scama.chunk_utilis import build_scama_mask_for_cross_attention_decoder
         self.build_scama_mask_for_cross_attention_decoder_fn = build_scama_mask_for_cross_attention_decoder
         self.decoder_attention_chunk_type = kwargs.get("decoder_attention_chunk_type", "chunk")
        self.scama_mask = None
        if hasattr(self.encoder, "overlap_chunk_cls") and self.encoder.overlap_chunk_cls is not None:
            from funasr.models.scama.chunk_utilis import build_scama_mask_for_cross_attention_decoder
            self.build_scama_mask_for_cross_attention_decoder_fn = build_scama_mask_for_cross_attention_decoder
            self.decoder_attention_chunk_type = kwargs.get("decoder_attention_chunk_type", "chunk")


	
   def forward(
      self,
      speech: torch.Tensor,
      speech_lengths: torch.Tensor,
      text: torch.Tensor,
      text_lengths: torch.Tensor,
      **kwargs,
   ) -> Tuple[torch.Tensor, Dict[str, torch.Tensor], torch.Tensor]:
      """Encoder + Decoder + Calc loss
      Args:
            speech: (Batch, Length, ...)
            speech_lengths: (Batch, )
            text: (Batch, Length)
            text_lengths: (Batch,)
      """
      # import pdb;
      # pdb.set_trace()
      decoding_ind = kwargs.get("decoding_ind")
      if len(text_lengths.size()) > 1:
         text_lengths = text_lengths[:, 0]
      if len(speech_lengths.size()) > 1:
         speech_lengths = speech_lengths[:, 0]
		
      batch_size = speech.shape[0]
		
      # Encoder
      if hasattr(self.encoder, "overlap_chunk_cls"):
         ind = self.encoder.overlap_chunk_cls.random_choice(self.training, decoding_ind)
         encoder_out, encoder_out_lens = self.encode(speech, speech_lengths, ind=ind)
      else:
         encoder_out, encoder_out_lens = self.encode(speech, speech_lengths)
		
      loss_ctc, cer_ctc = None, None
      loss_pre = None
      stats = dict()
		
      # decoder: CTC branch
    
    def forward(
        self,
        speech: torch.Tensor,
        speech_lengths: torch.Tensor,
        text: torch.Tensor,
        text_lengths: torch.Tensor,
        **kwargs,
    ) -> Tuple[torch.Tensor, Dict[str, torch.Tensor], torch.Tensor]:
        """Encoder + Decoder + Calc loss
        Args:
                speech: (Batch, Length, ...)
                speech_lengths: (Batch, )
                text: (Batch, Length)
                text_lengths: (Batch,)
        """
        # import pdb;
        # pdb.set_trace()
        decoding_ind = kwargs.get("decoding_ind")
        if len(text_lengths.size()) > 1:
            text_lengths = text_lengths[:, 0]
        if len(speech_lengths.size()) > 1:
            speech_lengths = speech_lengths[:, 0]
        
        batch_size = speech.shape[0]
        
        # Encoder
        if hasattr(self.encoder, "overlap_chunk_cls"):
            ind = self.encoder.overlap_chunk_cls.random_choice(self.training, decoding_ind)
            encoder_out, encoder_out_lens = self.encode(speech, speech_lengths, ind=ind)
        else:
            encoder_out, encoder_out_lens = self.encode(speech, speech_lengths)
        
        loss_ctc, cer_ctc = None, None
        loss_pre = None
        stats = dict()
        
        # decoder: CTC branch

      if self.ctc_weight > 0.0:
         if hasattr(self.encoder, "overlap_chunk_cls"):
            encoder_out_ctc, encoder_out_lens_ctc = self.encoder.overlap_chunk_cls.remove_chunk(encoder_out,
                                                                                                encoder_out_lens,
                                                                                                chunk_outs=None)
         else:
            encoder_out_ctc, encoder_out_lens_ctc = encoder_out, encoder_out_lens
				
         loss_ctc, cer_ctc = self._calc_ctc_loss(
            encoder_out_ctc, encoder_out_lens_ctc, text, text_lengths
         )
         # Collect CTC branch stats
         stats["loss_ctc"] = loss_ctc.detach() if loss_ctc is not None else None
         stats["cer_ctc"] = cer_ctc
		
      # decoder: Attention decoder branch
      loss_att, acc_att, cer_att, wer_att, loss_pre, pre_loss_att = self._calc_att_predictor_loss(
         encoder_out, encoder_out_lens, text, text_lengths
      )
		
      # 3. CTC-Att loss definition
      if self.ctc_weight == 0.0:
         loss = loss_att + loss_pre * self.predictor_weight
      else:
         loss = self.ctc_weight * loss_ctc + (
               1 - self.ctc_weight) * loss_att + loss_pre * self.predictor_weight
		
      # Collect Attn branch stats
      stats["loss_att"] = loss_att.detach() if loss_att is not None else None
      stats["pre_loss_att"] = pre_loss_att.detach() if pre_loss_att is not None else None
      stats["acc"] = acc_att
      stats["cer"] = cer_att
      stats["wer"] = wer_att
      stats["loss_pre"] = loss_pre.detach().cpu() if loss_pre is not None else None
		
      stats["loss"] = torch.clone(loss.detach())
		
      # force_gatherable: to-device and to-tensor if scalar for DataParallel
      if self.length_normalized_loss:
         batch_size = (text_lengths + self.predictor_bias).sum()
      loss, stats, weight = force_gatherable((loss, stats, batch_size), loss.device)
      return loss, stats, weight
	
   def encode_chunk(
      self, speech: torch.Tensor, speech_lengths: torch.Tensor, cache: dict = None, **kwargs,
   ) -> Tuple[torch.Tensor, torch.Tensor]:
      """Frontend + Encoder. Note that this method is used by asr_inference.py
      Args:
            speech: (Batch, Length, ...)
            speech_lengths: (Batch, )
            ind: int
      """
      with autocast(False):
			
         # Data augmentation
         if self.specaug is not None and self.training:
            speech, speech_lengths = self.specaug(speech, speech_lengths)
			
         # Normalization for feature: e.g. Global-CMVN, Utterance-CMVN
         if self.normalize is not None:
            speech, speech_lengths = self.normalize(speech, speech_lengths)
		
      # Forward encoder
      encoder_out, encoder_out_lens, _ = self.encoder.forward_chunk(speech, speech_lengths, cache=cache["encoder"])
      if isinstance(encoder_out, tuple):
         encoder_out = encoder_out[0]
		
      return encoder_out, torch.tensor([encoder_out.size(1)])
	
   def _calc_att_predictor_loss(
      self,
      encoder_out: torch.Tensor,
      encoder_out_lens: torch.Tensor,
      ys_pad: torch.Tensor,
      ys_pad_lens: torch.Tensor,
   ):
      encoder_out_mask = (~make_pad_mask(encoder_out_lens, maxlen=encoder_out.size(1))[:, None, :]).to(
         encoder_out.device)
      if self.predictor_bias == 1:
         _, ys_pad = add_sos_eos(ys_pad, self.sos, self.eos, self.ignore_id)
         ys_pad_lens = ys_pad_lens + self.predictor_bias
      mask_chunk_predictor = None
      if self.encoder.overlap_chunk_cls is not None:
         mask_chunk_predictor = self.encoder.overlap_chunk_cls.get_mask_chunk_predictor(None,
                                                                                        device=encoder_out.device,
                                                                                        batch_size=encoder_out.size(
                                                                                           0))
         mask_shfit_chunk = self.encoder.overlap_chunk_cls.get_mask_shfit_chunk(None, device=encoder_out.device,
                                                                                batch_size=encoder_out.size(0))
         encoder_out = encoder_out * mask_shfit_chunk
      pre_acoustic_embeds, pre_token_length, pre_alphas, _ = self.predictor(encoder_out,
                                                                            ys_pad,
                                                                            encoder_out_mask,
                                                                            ignore_id=self.ignore_id,
                                                                            mask_chunk_predictor=mask_chunk_predictor,
                                                                            target_label_length=ys_pad_lens,
                                                                            )
      predictor_alignments, predictor_alignments_len = self.predictor.gen_frame_alignments(pre_alphas,
                                                                                           encoder_out_lens)
		
      scama_mask = None
      if self.encoder.overlap_chunk_cls is not None and self.decoder_attention_chunk_type == 'chunk':
         encoder_chunk_size = self.encoder.overlap_chunk_cls.chunk_size_pad_shift_cur
         attention_chunk_center_bias = 0
         attention_chunk_size = encoder_chunk_size
         decoder_att_look_back_factor = self.encoder.overlap_chunk_cls.decoder_att_look_back_factor_cur
         mask_shift_att_chunk_decoder = self.encoder.overlap_chunk_cls. \
            get_mask_shift_att_chunk_decoder(None,
                                              device=encoder_out.device,
                                              batch_size=encoder_out.size(0)
                                              )
         scama_mask = self.build_scama_mask_for_cross_attention_decoder_fn(
            predictor_alignments=predictor_alignments,
            encoder_sequence_length=encoder_out_lens,
            chunk_size=1,
            encoder_chunk_size=encoder_chunk_size,
            attention_chunk_center_bias=attention_chunk_center_bias,
            attention_chunk_size=attention_chunk_size,
            attention_chunk_type=self.decoder_attention_chunk_type,
            step=None,
            predictor_mask_chunk_hopping=mask_chunk_predictor,
            decoder_att_look_back_factor=decoder_att_look_back_factor,
            mask_shift_att_chunk_decoder=mask_shift_att_chunk_decoder,
            target_length=ys_pad_lens,
            is_training=self.training,
         )
      elif self.encoder.overlap_chunk_cls is not None:
         encoder_out, encoder_out_lens = self.encoder.overlap_chunk_cls.remove_chunk(encoder_out,
                                                                                     encoder_out_lens,
                                                                                     chunk_outs=None)
      # 0. sampler
      decoder_out_1st = None
      pre_loss_att = None
      if self.sampling_ratio > 0.0:
         if self.step_cur < 2:
            logging.info("enable sampler in paraformer, sampling_ratio: {}".format(self.sampling_ratio))
         if self.use_1st_decoder_loss:
            sematic_embeds, decoder_out_1st, pre_loss_att = \
               self.sampler_with_grad(encoder_out, encoder_out_lens, ys_pad,
                                      ys_pad_lens, pre_acoustic_embeds, scama_mask)
         else:
            sematic_embeds, decoder_out_1st = \
               self.sampler(encoder_out, encoder_out_lens, ys_pad,
                            ys_pad_lens, pre_acoustic_embeds, scama_mask)
      else:
         if self.step_cur < 2:
            logging.info("disable sampler in paraformer, sampling_ratio: {}".format(self.sampling_ratio))
         sematic_embeds = pre_acoustic_embeds
		
      # 1. Forward decoder
      decoder_outs = self.decoder(
         encoder_out, encoder_out_lens, sematic_embeds, ys_pad_lens, scama_mask
      )
      decoder_out, _ = decoder_outs[0], decoder_outs[1]
		
      if decoder_out_1st is None:
         decoder_out_1st = decoder_out
      # 2. Compute attention loss
      loss_att = self.criterion_att(decoder_out, ys_pad)
      acc_att = th_accuracy(
         decoder_out_1st.view(-1, self.vocab_size),
         ys_pad,
         ignore_label=self.ignore_id,
      )
      loss_pre = self.criterion_pre(ys_pad_lens.type_as(pre_token_length), pre_token_length)
		
      # Compute cer/wer using attention-decoder
      if self.training or self.error_calculator is None:
         cer_att, wer_att = None, None
      else:
         ys_hat = decoder_out_1st.argmax(dim=-1)
         cer_att, wer_att = self.error_calculator(ys_hat.cpu(), ys_pad.cpu())
		
      return loss_att, acc_att, cer_att, wer_att, loss_pre, pre_loss_att
	
   def sampler(self, encoder_out, encoder_out_lens, ys_pad, ys_pad_lens, pre_acoustic_embeds, chunk_mask=None):
		
      tgt_mask = (~make_pad_mask(ys_pad_lens, maxlen=ys_pad_lens.max())[:, :, None]).to(ys_pad.device)
      ys_pad_masked = ys_pad * tgt_mask[:, :, 0]
      if self.share_embedding:
         ys_pad_embed = self.decoder.output_layer.weight[ys_pad_masked]
      else:
         ys_pad_embed = self.decoder.embed(ys_pad_masked)
      with torch.no_grad():
         decoder_outs = self.decoder(
            encoder_out, encoder_out_lens, pre_acoustic_embeds, ys_pad_lens, chunk_mask
         )
         decoder_out, _ = decoder_outs[0], decoder_outs[1]
         pred_tokens = decoder_out.argmax(-1)
         nonpad_positions = ys_pad.ne(self.ignore_id)
         seq_lens = (nonpad_positions).sum(1)
         same_num = ((pred_tokens == ys_pad) & nonpad_positions).sum(1)
         input_mask = torch.ones_like(nonpad_positions)
         bsz, seq_len = ys_pad.size()
         for li in range(bsz):
            target_num = (((seq_lens[li] - same_num[li].sum()).float()) * self.sampling_ratio).long()
            if target_num > 0:
               input_mask[li].scatter_(dim=0, index=torch.randperm(seq_lens[li])[:target_num].cuda(), value=0)
         input_mask = input_mask.eq(1)
         input_mask = input_mask.masked_fill(~nonpad_positions, False)
         input_mask_expand_dim = input_mask.unsqueeze(2).to(pre_acoustic_embeds.device)
		
      sematic_embeds = pre_acoustic_embeds.masked_fill(~input_mask_expand_dim, 0) + ys_pad_embed.masked_fill(
         input_mask_expand_dim, 0)
      return sematic_embeds * tgt_mask, decoder_out * tgt_mask
	
        if self.ctc_weight > 0.0:
            if hasattr(self.encoder, "overlap_chunk_cls"):
                encoder_out_ctc, encoder_out_lens_ctc = self.encoder.overlap_chunk_cls.remove_chunk(encoder_out,
                                                                                                    encoder_out_lens,
                                                                                                    chunk_outs=None)
            else:
                encoder_out_ctc, encoder_out_lens_ctc = encoder_out, encoder_out_lens
                
            loss_ctc, cer_ctc = self._calc_ctc_loss(
                encoder_out_ctc, encoder_out_lens_ctc, text, text_lengths
            )
            # Collect CTC branch stats
            stats["loss_ctc"] = loss_ctc.detach() if loss_ctc is not None else None
            stats["cer_ctc"] = cer_ctc
        
        # decoder: Attention decoder branch
        loss_att, acc_att, cer_att, wer_att, loss_pre, pre_loss_att = self._calc_att_predictor_loss(
            encoder_out, encoder_out_lens, text, text_lengths
        )
        
        # 3. CTC-Att loss definition
        if self.ctc_weight == 0.0:
            loss = loss_att + loss_pre * self.predictor_weight
        else:
            loss = self.ctc_weight * loss_ctc + (
                    1 - self.ctc_weight) * loss_att + loss_pre * self.predictor_weight
        
        # Collect Attn branch stats
        stats["loss_att"] = loss_att.detach() if loss_att is not None else None
        stats["pre_loss_att"] = pre_loss_att.detach() if pre_loss_att is not None else None
        stats["acc"] = acc_att
        stats["cer"] = cer_att
        stats["wer"] = wer_att
        stats["loss_pre"] = loss_pre.detach().cpu() if loss_pre is not None else None
        
        stats["loss"] = torch.clone(loss.detach())
        
        # force_gatherable: to-device and to-tensor if scalar for DataParallel
        if self.length_normalized_loss:
            batch_size = (text_lengths + self.predictor_bias).sum()
        loss, stats, weight = force_gatherable((loss, stats, batch_size), loss.device)
        return loss, stats, weight
    
    def encode_chunk(
        self, speech: torch.Tensor, speech_lengths: torch.Tensor, cache: dict = None, **kwargs,
    ) -> Tuple[torch.Tensor, torch.Tensor]:
        """Frontend + Encoder. Note that this method is used by asr_inference.py
        Args:
                speech: (Batch, Length, ...)
                speech_lengths: (Batch, )
                ind: int
        """
        with autocast(False):
            
            # Data augmentation
            if self.specaug is not None and self.training:
                speech, speech_lengths = self.specaug(speech, speech_lengths)
            
            # Normalization for feature: e.g. Global-CMVN, Utterance-CMVN
            if self.normalize is not None:
                speech, speech_lengths = self.normalize(speech, speech_lengths)
        
        # Forward encoder
        encoder_out, encoder_out_lens, _ = self.encoder.forward_chunk(speech, speech_lengths, cache=cache["encoder"])
        if isinstance(encoder_out, tuple):
            encoder_out = encoder_out[0]
        
        return encoder_out, torch.tensor([encoder_out.size(1)])
    
    def _calc_att_predictor_loss(
        self,
        encoder_out: torch.Tensor,
        encoder_out_lens: torch.Tensor,
        ys_pad: torch.Tensor,
        ys_pad_lens: torch.Tensor,
    ):
        encoder_out_mask = (~make_pad_mask(encoder_out_lens, maxlen=encoder_out.size(1))[:, None, :]).to(
            encoder_out.device)
        if self.predictor_bias == 1:
            _, ys_pad = add_sos_eos(ys_pad, self.sos, self.eos, self.ignore_id)
            ys_pad_lens = ys_pad_lens + self.predictor_bias
        mask_chunk_predictor = None
        if self.encoder.overlap_chunk_cls is not None:
            mask_chunk_predictor = self.encoder.overlap_chunk_cls.get_mask_chunk_predictor(None,
                                                                                           device=encoder_out.device,
                                                                                           batch_size=encoder_out.size(
                                                                                               0))
            mask_shfit_chunk = self.encoder.overlap_chunk_cls.get_mask_shfit_chunk(None, device=encoder_out.device,
                                                                                   batch_size=encoder_out.size(0))
            encoder_out = encoder_out * mask_shfit_chunk
        pre_acoustic_embeds, pre_token_length, pre_alphas, _ = self.predictor(encoder_out,
                                                                              ys_pad,
                                                                              encoder_out_mask,
                                                                              ignore_id=self.ignore_id,
                                                                              mask_chunk_predictor=mask_chunk_predictor,
                                                                              target_label_length=ys_pad_lens,
                                                                              )
        predictor_alignments, predictor_alignments_len = self.predictor.gen_frame_alignments(pre_alphas,
                                                                                             encoder_out_lens)
        
        scama_mask = None
        if self.encoder.overlap_chunk_cls is not None and self.decoder_attention_chunk_type == 'chunk':
            encoder_chunk_size = self.encoder.overlap_chunk_cls.chunk_size_pad_shift_cur
            attention_chunk_center_bias = 0
            attention_chunk_size = encoder_chunk_size
            decoder_att_look_back_factor = self.encoder.overlap_chunk_cls.decoder_att_look_back_factor_cur
            mask_shift_att_chunk_decoder = self.encoder.overlap_chunk_cls. \
                get_mask_shift_att_chunk_decoder(None,
                                                 device=encoder_out.device,
                                                 batch_size=encoder_out.size(0)
                                                 )
            scama_mask = self.build_scama_mask_for_cross_attention_decoder_fn(
                predictor_alignments=predictor_alignments,
                encoder_sequence_length=encoder_out_lens,
                chunk_size=1,
                encoder_chunk_size=encoder_chunk_size,
                attention_chunk_center_bias=attention_chunk_center_bias,
                attention_chunk_size=attention_chunk_size,
                attention_chunk_type=self.decoder_attention_chunk_type,
                step=None,
                predictor_mask_chunk_hopping=mask_chunk_predictor,
                decoder_att_look_back_factor=decoder_att_look_back_factor,
                mask_shift_att_chunk_decoder=mask_shift_att_chunk_decoder,
                target_length=ys_pad_lens,
                is_training=self.training,
            )
        elif self.encoder.overlap_chunk_cls is not None:
            encoder_out, encoder_out_lens = self.encoder.overlap_chunk_cls.remove_chunk(encoder_out,
                                                                                        encoder_out_lens,
                                                                                        chunk_outs=None)
        # 0. sampler
        decoder_out_1st = None
        pre_loss_att = None
        if self.sampling_ratio > 0.0:
            if self.step_cur < 2:
                logging.info("enable sampler in paraformer, sampling_ratio: {}".format(self.sampling_ratio))
            if self.use_1st_decoder_loss:
                sematic_embeds, decoder_out_1st, pre_loss_att = \
                    self.sampler_with_grad(encoder_out, encoder_out_lens, ys_pad,
                                           ys_pad_lens, pre_acoustic_embeds, scama_mask)
            else:
                sematic_embeds, decoder_out_1st = \
                    self.sampler(encoder_out, encoder_out_lens, ys_pad,
                                 ys_pad_lens, pre_acoustic_embeds, scama_mask)
        else:
            if self.step_cur < 2:
                logging.info("disable sampler in paraformer, sampling_ratio: {}".format(self.sampling_ratio))
            sematic_embeds = pre_acoustic_embeds
        
        # 1. Forward decoder
        decoder_outs = self.decoder(
            encoder_out, encoder_out_lens, sematic_embeds, ys_pad_lens, scama_mask
        )
        decoder_out, _ = decoder_outs[0], decoder_outs[1]
        
        if decoder_out_1st is None:
            decoder_out_1st = decoder_out
        # 2. Compute attention loss
        loss_att = self.criterion_att(decoder_out, ys_pad)
        acc_att = th_accuracy(
            decoder_out_1st.view(-1, self.vocab_size),
            ys_pad,
            ignore_label=self.ignore_id,
        )
        loss_pre = self.criterion_pre(ys_pad_lens.type_as(pre_token_length), pre_token_length)
        
        # Compute cer/wer using attention-decoder
        if self.training or self.error_calculator is None:
            cer_att, wer_att = None, None
        else:
            ys_hat = decoder_out_1st.argmax(dim=-1)
            cer_att, wer_att = self.error_calculator(ys_hat.cpu(), ys_pad.cpu())
        
        return loss_att, acc_att, cer_att, wer_att, loss_pre, pre_loss_att
    
    def sampler(self, encoder_out, encoder_out_lens, ys_pad, ys_pad_lens, pre_acoustic_embeds, chunk_mask=None):
        
        tgt_mask = (~make_pad_mask(ys_pad_lens, maxlen=ys_pad_lens.max())[:, :, None]).to(ys_pad.device)
        ys_pad_masked = ys_pad * tgt_mask[:, :, 0]
        if self.share_embedding:
            ys_pad_embed = self.decoder.output_layer.weight[ys_pad_masked]
        else:
            ys_pad_embed = self.decoder.embed(ys_pad_masked)
        with torch.no_grad():
            decoder_outs = self.decoder(
                encoder_out, encoder_out_lens, pre_acoustic_embeds, ys_pad_lens, chunk_mask
            )
            decoder_out, _ = decoder_outs[0], decoder_outs[1]
            pred_tokens = decoder_out.argmax(-1)
            nonpad_positions = ys_pad.ne(self.ignore_id)
            seq_lens = (nonpad_positions).sum(1)
            same_num = ((pred_tokens == ys_pad) & nonpad_positions).sum(1)
            input_mask = torch.ones_like(nonpad_positions)
            bsz, seq_len = ys_pad.size()
            for li in range(bsz):
                target_num = (((seq_lens[li] - same_num[li].sum()).float()) * self.sampling_ratio).long()
                if target_num > 0:
                    input_mask[li].scatter_(dim=0, index=torch.randperm(seq_lens[li])[:target_num].cuda(), value=0)
            input_mask = input_mask.eq(1)
            input_mask = input_mask.masked_fill(~nonpad_positions, False)
            input_mask_expand_dim = input_mask.unsqueeze(2).to(pre_acoustic_embeds.device)
        
        sematic_embeds = pre_acoustic_embeds.masked_fill(~input_mask_expand_dim, 0) + ys_pad_embed.masked_fill(
            input_mask_expand_dim, 0)
        return sematic_embeds * tgt_mask, decoder_out * tgt_mask
    

   def calc_predictor(self, encoder_out, encoder_out_lens):
		
      encoder_out_mask = (~make_pad_mask(encoder_out_lens, maxlen=encoder_out.size(1))[:, None, :]).to(
         encoder_out.device)
      mask_chunk_predictor = None
      if self.encoder.overlap_chunk_cls is not None:
         mask_chunk_predictor = self.encoder.overlap_chunk_cls.get_mask_chunk_predictor(None,
                                                                                        device=encoder_out.device,
                                                                                        batch_size=encoder_out.size(
                                                                                           0))
         mask_shfit_chunk = self.encoder.overlap_chunk_cls.get_mask_shfit_chunk(None, device=encoder_out.device,
                                                                                batch_size=encoder_out.size(0))
         encoder_out = encoder_out * mask_shfit_chunk
      pre_acoustic_embeds, pre_token_length, pre_alphas, pre_peak_index = self.predictor(encoder_out,
                                                                                         None,
                                                                                         encoder_out_mask,
                                                                                         ignore_id=self.ignore_id,
                                                                                         mask_chunk_predictor=mask_chunk_predictor,
                                                                                         target_label_length=None,
                                                                                         )
      predictor_alignments, predictor_alignments_len = self.predictor.gen_frame_alignments(pre_alphas,
                                                                                           encoder_out_lens + 1 if self.predictor.tail_threshold > 0.0 else encoder_out_lens)
		
      scama_mask = None
      if self.encoder.overlap_chunk_cls is not None and self.decoder_attention_chunk_type == 'chunk':
         encoder_chunk_size = self.encoder.overlap_chunk_cls.chunk_size_pad_shift_cur
         attention_chunk_center_bias = 0
         attention_chunk_size = encoder_chunk_size
         decoder_att_look_back_factor = self.encoder.overlap_chunk_cls.decoder_att_look_back_factor_cur
         mask_shift_att_chunk_decoder = self.encoder.overlap_chunk_cls. \
            get_mask_shift_att_chunk_decoder(None,
                                              device=encoder_out.device,
                                              batch_size=encoder_out.size(0)
                                              )
         scama_mask = self.build_scama_mask_for_cross_attention_decoder_fn(
            predictor_alignments=predictor_alignments,
            encoder_sequence_length=encoder_out_lens,
            chunk_size=1,
            encoder_chunk_size=encoder_chunk_size,
            attention_chunk_center_bias=attention_chunk_center_bias,
            attention_chunk_size=attention_chunk_size,
            attention_chunk_type=self.decoder_attention_chunk_type,
            step=None,
            predictor_mask_chunk_hopping=mask_chunk_predictor,
            decoder_att_look_back_factor=decoder_att_look_back_factor,
            mask_shift_att_chunk_decoder=mask_shift_att_chunk_decoder,
            target_length=None,
            is_training=self.training,
         )
      self.scama_mask = scama_mask
		
      return pre_acoustic_embeds, pre_token_length, pre_alphas, pre_peak_index
	
   def calc_predictor_chunk(self, encoder_out, encoder_out_lens, cache=None, **kwargs):
      is_final = kwargs.get("is_final", False)
    def calc_predictor(self, encoder_out, encoder_out_lens):
        
        encoder_out_mask = (~make_pad_mask(encoder_out_lens, maxlen=encoder_out.size(1))[:, None, :]).to(
            encoder_out.device)
        mask_chunk_predictor = None
        if self.encoder.overlap_chunk_cls is not None:
            mask_chunk_predictor = self.encoder.overlap_chunk_cls.get_mask_chunk_predictor(None,
                                                                                           device=encoder_out.device,
                                                                                           batch_size=encoder_out.size(
                                                                                               0))
            mask_shfit_chunk = self.encoder.overlap_chunk_cls.get_mask_shfit_chunk(None, device=encoder_out.device,
                                                                                   batch_size=encoder_out.size(0))
            encoder_out = encoder_out * mask_shfit_chunk
        pre_acoustic_embeds, pre_token_length, pre_alphas, pre_peak_index = self.predictor(encoder_out,
                                                                                           None,
                                                                                           encoder_out_mask,
                                                                                           ignore_id=self.ignore_id,
                                                                                           mask_chunk_predictor=mask_chunk_predictor,
                                                                                           target_label_length=None,
                                                                                           )
        predictor_alignments, predictor_alignments_len = self.predictor.gen_frame_alignments(pre_alphas,
                                                                                             encoder_out_lens + 1 if self.predictor.tail_threshold > 0.0 else encoder_out_lens)
        
        scama_mask = None
        if self.encoder.overlap_chunk_cls is not None and self.decoder_attention_chunk_type == 'chunk':
            encoder_chunk_size = self.encoder.overlap_chunk_cls.chunk_size_pad_shift_cur
            attention_chunk_center_bias = 0
            attention_chunk_size = encoder_chunk_size
            decoder_att_look_back_factor = self.encoder.overlap_chunk_cls.decoder_att_look_back_factor_cur
            mask_shift_att_chunk_decoder = self.encoder.overlap_chunk_cls. \
                get_mask_shift_att_chunk_decoder(None,
                                                 device=encoder_out.device,
                                                 batch_size=encoder_out.size(0)
                                                 )
            scama_mask = self.build_scama_mask_for_cross_attention_decoder_fn(
                predictor_alignments=predictor_alignments,
                encoder_sequence_length=encoder_out_lens,
                chunk_size=1,
                encoder_chunk_size=encoder_chunk_size,
                attention_chunk_center_bias=attention_chunk_center_bias,
                attention_chunk_size=attention_chunk_size,
                attention_chunk_type=self.decoder_attention_chunk_type,
                step=None,
                predictor_mask_chunk_hopping=mask_chunk_predictor,
                decoder_att_look_back_factor=decoder_att_look_back_factor,
                mask_shift_att_chunk_decoder=mask_shift_att_chunk_decoder,
                target_length=None,
                is_training=self.training,
            )
        self.scama_mask = scama_mask
        
        return pre_acoustic_embeds, pre_token_length, pre_alphas, pre_peak_index
    
    def calc_predictor_chunk(self, encoder_out, encoder_out_lens, cache=None, **kwargs):
        is_final = kwargs.get("is_final", False)

      return self.predictor.forward_chunk(encoder_out, cache["encoder"], is_final=is_final)
	
   def cal_decoder_with_predictor(self, encoder_out, encoder_out_lens, sematic_embeds, ys_pad_lens):
      decoder_outs = self.decoder(
         encoder_out, encoder_out_lens, sematic_embeds, ys_pad_lens, self.scama_mask
      )
      decoder_out = decoder_outs[0]
      decoder_out = torch.log_softmax(decoder_out, dim=-1)
      return decoder_out, ys_pad_lens
	
   def cal_decoder_with_predictor_chunk(self, encoder_out, encoder_out_lens, sematic_embeds, ys_pad_lens, cache=None):
      decoder_outs = self.decoder.forward_chunk(
         encoder_out, sematic_embeds, cache["decoder"]
      )
      decoder_out = decoder_outs
      decoder_out = torch.log_softmax(decoder_out, dim=-1)
      return decoder_out, ys_pad_lens
	
   def init_cache(self, cache: dict = {}, **kwargs):
      chunk_size = kwargs.get("chunk_size", [0, 10, 5])
      encoder_chunk_look_back = kwargs.get("encoder_chunk_look_back", 0)
      decoder_chunk_look_back = kwargs.get("decoder_chunk_look_back", 0)
      batch_size = 1
        return self.predictor.forward_chunk(encoder_out, cache["encoder"], is_final=is_final)
    
    def cal_decoder_with_predictor(self, encoder_out, encoder_out_lens, sematic_embeds, ys_pad_lens):
        decoder_outs = self.decoder(
            encoder_out, encoder_out_lens, sematic_embeds, ys_pad_lens, self.scama_mask
        )
        decoder_out = decoder_outs[0]
        decoder_out = torch.log_softmax(decoder_out, dim=-1)
        return decoder_out, ys_pad_lens
    
    def cal_decoder_with_predictor_chunk(self, encoder_out, encoder_out_lens, sematic_embeds, ys_pad_lens, cache=None):
        decoder_outs = self.decoder.forward_chunk(
            encoder_out, sematic_embeds, cache["decoder"]
        )
        decoder_out = decoder_outs
        decoder_out = torch.log_softmax(decoder_out, dim=-1)
        return decoder_out, ys_pad_lens
    
    def init_cache(self, cache: dict = {}, **kwargs):
        chunk_size = kwargs.get("chunk_size", [0, 10, 5])
        encoder_chunk_look_back = kwargs.get("encoder_chunk_look_back", 0)
        decoder_chunk_look_back = kwargs.get("decoder_chunk_look_back", 0)
        batch_size = 1

      enc_output_size = kwargs["encoder_conf"]["output_size"]
      feats_dims = kwargs["frontend_conf"]["n_mels"] * kwargs["frontend_conf"]["lfr_m"]
      cache_encoder = {"start_idx": 0, "cif_hidden": torch.zeros((batch_size, 1, enc_output_size)),
                  "cif_alphas": torch.zeros((batch_size, 1)), "chunk_size": chunk_size,
                  "encoder_chunk_look_back": encoder_chunk_look_back, "last_chunk": False, "opt": None,
                  "feats": torch.zeros((batch_size, chunk_size[0] + chunk_size[2], feats_dims)),
                  "tail_chunk": False}
      cache["encoder"] = cache_encoder
		
      cache_decoder = {"decode_fsmn": None, "decoder_chunk_look_back": decoder_chunk_look_back, "opt": None,
                  "chunk_size": chunk_size}
      cache["decoder"] = cache_decoder
      cache["frontend"] = {}
      cache["prev_samples"] = torch.empty(0)
		
      return cache
	
   def generate_chunk(self,
                      speech,
                      speech_lengths=None,
                      key: list = None,
                      tokenizer=None,
                      frontend=None,
                      **kwargs,
                      ):
      cache = kwargs.get("cache", {})
      speech.to(device=kwargs["device"]), speech_lengths.to(device=kwargs["device"])
		
      # Encoder
      encoder_out, encoder_out_lens = self.encode_chunk(speech, speech_lengths, cache=cache, is_final=kwargs.get("is_final", False))
      if isinstance(encoder_out, tuple):
         encoder_out = encoder_out[0]
		
      # predictor
      predictor_outs = self.calc_predictor_chunk(encoder_out, encoder_out_lens, cache=cache, is_final=kwargs.get("is_final", False))
      pre_acoustic_embeds, pre_token_length, alphas, pre_peak_index = predictor_outs[0], predictor_outs[1], \
                                                                      predictor_outs[2], predictor_outs[3]
      pre_token_length = pre_token_length.round().long()
      if torch.max(pre_token_length) < 1:
         return []
      decoder_outs = self.cal_decoder_with_predictor_chunk(encoder_out,
                                                           encoder_out_lens,
                                                           pre_acoustic_embeds,
                                                           pre_token_length,
                                                           cache=cache
                                                           )
      decoder_out, ys_pad_lens = decoder_outs[0], decoder_outs[1]
        enc_output_size = kwargs["encoder_conf"]["output_size"]
        feats_dims = kwargs["frontend_conf"]["n_mels"] * kwargs["frontend_conf"]["lfr_m"]
        cache_encoder = {"start_idx": 0, "cif_hidden": torch.zeros((batch_size, 1, enc_output_size)),
                    "cif_alphas": torch.zeros((batch_size, 1)), "chunk_size": chunk_size,
                    "encoder_chunk_look_back": encoder_chunk_look_back, "last_chunk": False, "opt": None,
                    "feats": torch.zeros((batch_size, chunk_size[0] + chunk_size[2], feats_dims)),
                    "tail_chunk": False}
        cache["encoder"] = cache_encoder
        
        cache_decoder = {"decode_fsmn": None, "decoder_chunk_look_back": decoder_chunk_look_back, "opt": None,
                    "chunk_size": chunk_size}
        cache["decoder"] = cache_decoder
        cache["frontend"] = {}
        cache["prev_samples"] = torch.empty(0)
        
        return cache
    
    def generate_chunk(self,
                       speech,
                       speech_lengths=None,
                       key: list = None,
                       tokenizer=None,
                       frontend=None,
                       **kwargs,
                       ):
        cache = kwargs.get("cache", {})
        speech = speech.to(device=kwargs["device"])
        speech_lengths = speech_lengths.to(device=kwargs["device"])
        
        # Encoder
        encoder_out, encoder_out_lens = self.encode_chunk(speech, speech_lengths, cache=cache, is_final=kwargs.get("is_final", False))
        if isinstance(encoder_out, tuple):
            encoder_out = encoder_out[0]
        
        # predictor
        predictor_outs = self.calc_predictor_chunk(encoder_out, encoder_out_lens, cache=cache, is_final=kwargs.get("is_final", False))
        pre_acoustic_embeds, pre_token_length, alphas, pre_peak_index = predictor_outs[0], predictor_outs[1], \
                                                                        predictor_outs[2], predictor_outs[3]
        pre_token_length = pre_token_length.round().long()
        if torch.max(pre_token_length) < 1:
            return []
        decoder_outs = self.cal_decoder_with_predictor_chunk(encoder_out,
                                                             encoder_out_lens,
                                                             pre_acoustic_embeds,
                                                             pre_token_length,
                                                             cache=cache
                                                             )
        decoder_out, ys_pad_lens = decoder_outs[0], decoder_outs[1]

      results = []
      b, n, d = decoder_out.size()
      if isinstance(key[0], (list, tuple)):
         key = key[0]
      for i in range(b):
         x = encoder_out[i, :encoder_out_lens[i], :]
         am_scores = decoder_out[i, :pre_token_length[i], :]
         if self.beam_search is not None:
            nbest_hyps = self.beam_search(
               x=x, am_scores=am_scores, maxlenratio=kwargs.get("maxlenratio", 0.0),
               minlenratio=kwargs.get("minlenratio", 0.0)
            )
				
            nbest_hyps = nbest_hyps[: self.nbest]
         else:
				
            yseq = am_scores.argmax(dim=-1)
            score = am_scores.max(dim=-1)[0]
            score = torch.sum(score, dim=-1)
            # pad with mask tokens to ensure compatibility with sos/eos tokens
            yseq = torch.tensor(
               [self.sos] + yseq.tolist() + [self.eos], device=yseq.device
            )
            nbest_hyps = [Hypothesis(yseq=yseq, score=score)]
         for nbest_idx, hyp in enumerate(nbest_hyps):
				
            # remove sos/eos and get results
            last_pos = -1
            if isinstance(hyp.yseq, list):
               token_int = hyp.yseq[1:last_pos]
            else:
               token_int = hyp.yseq[1:last_pos].tolist()
				
            # remove blank symbol id, which is assumed to be 0
            token_int = list(filter(lambda x: x != self.eos and x != self.sos and x != self.blank_id, token_int))
				
        results = []
        b, n, d = decoder_out.size()
        if isinstance(key[0], (list, tuple)):
            key = key[0]
        for i in range(b):
            x = encoder_out[i, :encoder_out_lens[i], :]
            am_scores = decoder_out[i, :pre_token_length[i], :]
            if self.beam_search is not None:
                nbest_hyps = self.beam_search(
                    x=x, am_scores=am_scores, maxlenratio=kwargs.get("maxlenratio", 0.0),
                    minlenratio=kwargs.get("minlenratio", 0.0)
                )
                
                nbest_hyps = nbest_hyps[: self.nbest]
            else:
                
                yseq = am_scores.argmax(dim=-1)
                score = am_scores.max(dim=-1)[0]
                score = torch.sum(score, dim=-1)
                # pad with mask tokens to ensure compatibility with sos/eos tokens
                yseq = torch.tensor(
                    [self.sos] + yseq.tolist() + [self.eos], device=yseq.device
                )
                nbest_hyps = [Hypothesis(yseq=yseq, score=score)]
            for nbest_idx, hyp in enumerate(nbest_hyps):
                
                # remove sos/eos and get results
                last_pos = -1
                if isinstance(hyp.yseq, list):
                    token_int = hyp.yseq[1:last_pos]
                else:
                    token_int = hyp.yseq[1:last_pos].tolist()
                
                # remove blank symbol id, which is assumed to be 0
                token_int = list(filter(lambda x: x != self.eos and x != self.sos and x != self.blank_id, token_int))
                

            # Change integer-ids to tokens
            token = tokenizer.ids2tokens(token_int)
            # text = tokenizer.tokens2text(token)
				
            result_i = token
                # Change integer-ids to tokens
                token = tokenizer.ids2tokens(token_int)
                # text = tokenizer.tokens2text(token)
                
                result_i = token


            results.extend(result_i)
		
      return results
	
   def generate(self,
                data_in,
                data_lengths=None,
                key: list = None,
                tokenizer=None,
                frontend=None,
                cache: dict={},
                **kwargs,
                ):
                results.extend(result_i)
        
        return results
    
    def generate(self,
                 data_in,
                 data_lengths=None,
                 key: list = None,
                 tokenizer=None,
                 frontend=None,
                 cache: dict={},
                 **kwargs,
                 ):

      # init beamsearch
      is_use_ctc = kwargs.get("decoding_ctc_weight", 0.0) > 0.00001 and self.ctc != None
      is_use_lm = kwargs.get("lm_weight", 0.0) > 0.00001 and kwargs.get("lm_file", None) is not None
      if self.beam_search is None and (is_use_lm or is_use_ctc):
         logging.info("enable beam_search")
         self.init_beam_search(**kwargs)
         self.nbest = kwargs.get("nbest", 1)
		
        # init beamsearch
        is_use_ctc = kwargs.get("decoding_ctc_weight", 0.0) > 0.00001 and self.ctc != None
        is_use_lm = kwargs.get("lm_weight", 0.0) > 0.00001 and kwargs.get("lm_file", None) is not None
        if self.beam_search is None and (is_use_lm or is_use_ctc):
            logging.info("enable beam_search")
            self.init_beam_search(**kwargs)
            self.nbest = kwargs.get("nbest", 1)
        

      if len(cache) == 0:
         self.init_cache(cache, **kwargs)
		
		
      meta_data = {}
      chunk_size = kwargs.get("chunk_size", [0, 10, 5])
      chunk_stride_samples = int(chunk_size[1] * 960)  # 600ms
		
      time1 = time.perf_counter()
      cfg = {"is_final": kwargs.get("is_final", False)}
      audio_sample_list = load_audio_text_image_video(data_in,
                                          fs=frontend.fs,
                                          audio_fs=kwargs.get("fs", 16000),
                                          data_type=kwargs.get("data_type", "sound"),
                                          tokenizer=tokenizer,
                                          cache=cfg,
                                          )
      _is_final = cfg["is_final"] # if data_in is a file or url, set is_final=True
		
      time2 = time.perf_counter()
      meta_data["load_data"] = f"{time2 - time1:0.3f}"
      assert len(audio_sample_list) == 1, "batch_size must be set 1"
		
      audio_sample = torch.cat((cache["prev_samples"], audio_sample_list[0]))
		
      n = int(len(audio_sample) // chunk_stride_samples + int(_is_final))
      m = int(len(audio_sample) % chunk_stride_samples * (1-int(_is_final)))
      tokens = []
      for i in range(n):
         kwargs["is_final"] = _is_final and i == n -1
         audio_sample_i = audio_sample[i*chunk_stride_samples:(i+1)*chunk_stride_samples]
        if len(cache) == 0:
            self.init_cache(cache, **kwargs)
        
        
        meta_data = {}
        chunk_size = kwargs.get("chunk_size", [0, 10, 5])
        chunk_stride_samples = int(chunk_size[1] * 960)  # 600ms
        
        time1 = time.perf_counter()
        cfg = {"is_final": kwargs.get("is_final", False)}
        audio_sample_list = load_audio_text_image_video(data_in,
                                                        fs=frontend.fs,
                                                        audio_fs=kwargs.get("fs", 16000),
                                                        data_type=kwargs.get("data_type", "sound"),
                                                        tokenizer=tokenizer,
                                                        cache=cfg,
                                                        )
        _is_final = cfg["is_final"] # if data_in is a file or url, set is_final=True
        
        time2 = time.perf_counter()
        meta_data["load_data"] = f"{time2 - time1:0.3f}"
        assert len(audio_sample_list) == 1, "batch_size must be set 1"
        
        audio_sample = torch.cat((cache["prev_samples"], audio_sample_list[0]))
        
        n = int(len(audio_sample) // chunk_stride_samples + int(_is_final))
        m = int(len(audio_sample) % chunk_stride_samples * (1-int(_is_final)))
        tokens = []
        for i in range(n):
            kwargs["is_final"] = _is_final and i == n -1
            audio_sample_i = audio_sample[i*chunk_stride_samples:(i+1)*chunk_stride_samples]

         # extract fbank feats
         speech, speech_lengths = extract_fbank([audio_sample_i], data_type=kwargs.get("data_type", "sound"),
                                                frontend=frontend, cache=cache["frontend"], is_final=kwargs["is_final"])
         time3 = time.perf_counter()
         meta_data["extract_feat"] = f"{time3 - time2:0.3f}"
         meta_data["batch_data_time"] = speech_lengths.sum().item() * frontend.frame_shift * frontend.lfr_n / 1000
			
         tokens_i = self.generate_chunk(speech, speech_lengths, key=key, tokenizer=tokenizer, cache=cache, frontend=frontend, **kwargs)
         tokens.extend(tokens_i)
			
      text_postprocessed, _ = postprocess_utils.sentence_postprocess(tokens)
		
      result_i = {"key": key[0], "text": text_postprocessed}
      result = [result_i]
		
		
      cache["prev_samples"] = audio_sample[:-m]
      if _is_final:
         self.init_cache(cache, **kwargs)
		
      if kwargs.get("output_dir"):
         writer = DatadirWriter(kwargs.get("output_dir"))
         ibest_writer = writer[f"{1}best_recog"]
         ibest_writer["token"][key[0]] = " ".join(tokens)
         ibest_writer["text"][key[0]] = text_postprocessed
		
      return result, meta_data
            # extract fbank feats
            speech, speech_lengths = extract_fbank([audio_sample_i], data_type=kwargs.get("data_type", "sound"),
                                                   frontend=frontend, cache=cache["frontend"], is_final=kwargs["is_final"])
            time3 = time.perf_counter()
            meta_data["extract_feat"] = f"{time3 - time2:0.3f}"
            meta_data["batch_data_time"] = speech_lengths.sum().item() * frontend.frame_shift * frontend.lfr_n / 1000
            
            tokens_i = self.generate_chunk(speech, speech_lengths, key=key, tokenizer=tokenizer, cache=cache, frontend=frontend, **kwargs)
            tokens.extend(tokens_i)
            
        text_postprocessed, _ = postprocess_utils.sentence_postprocess(tokens)
        
        result_i = {"key": key[0], "text": text_postprocessed}
        result = [result_i]
        
        
        cache["prev_samples"] = audio_sample[:-m]
        if _is_final:
            self.init_cache(cache, **kwargs)
        
        if kwargs.get("output_dir"):
            writer = DatadirWriter(kwargs.get("output_dir"))
            ibest_writer = writer[f"{1}best_recog"]
            ibest_writer["token"][key[0]] = " ".join(tokens)
            ibest_writer["text"][key[0]] = text_postprocessed
        
        return result, meta_data