From aa910b9860d420877d73f36c71302995587b0a49 Mon Sep 17 00:00:00 2001
From: shixian.shi <shixian.shi@alibaba-inc.com>
Date: 星期四, 27 四月 2023 12:03:31 +0800
Subject: [PATCH] update adavanced clas, including model and dataset
---
funasr/datasets/large_datasets/dataset.py | 40 ++++
funasr/models/e2e_asr_contextual_paraformer.py | 408 +++++++++++++++++++++++++++++++++++++++++++++
funasr/tasks/asr.py | 2
funasr/datasets/large_datasets/utils/hotword_utils.py | 32 +++
funasr/datasets/large_datasets/utils/padding.py | 43 ++++
funasr/datasets/large_datasets/utils/tokenize.py | 7
6 files changed, 528 insertions(+), 4 deletions(-)
diff --git a/funasr/datasets/large_datasets/dataset.py b/funasr/datasets/large_datasets/dataset.py
index b0e1b8f..500257c 100644
--- a/funasr/datasets/large_datasets/dataset.py
+++ b/funasr/datasets/large_datasets/dataset.py
@@ -28,7 +28,7 @@
class AudioDataset(IterableDataset):
- def __init__(self, scp_lists, data_names, data_types, frontend_conf=None, shuffle=True, mode="train"):
+ def __init__(self, scp_lists, data_names, data_types, frontend_conf=None, shuffle=True, mode="train", pre_hwlist=None, pre_prob=0.0):
self.scp_lists = scp_lists
self.data_names = data_names
self.data_types = data_types
@@ -40,6 +40,8 @@
self.world_size = 1
self.worker_id = 0
self.num_workers = 1
+ self.pre_hwlist = pre_hwlist
+ self.pre_prob = pre_prob
def set_epoch(self, epoch):
self.epoch = epoch
@@ -131,6 +133,13 @@
sample_dict["sampling_rate"] = sampling_rate
if data_name == "speech":
sample_dict["key"] = key
+ elif data_type == "text_hotword":
+ text = item
+ segs = text.strip().split()
+ sample_dict[data_name] = segs[1:]
+ if "key" not in sample_dict:
+ sample_dict["key"] = segs[0]
+ sample_dict['hw_tag'] = 1
else:
text = item
segs = text.strip().split()
@@ -167,14 +176,39 @@
shuffle = conf.get('shuffle', True)
data_names = conf.get("data_names", "speech,text")
data_types = conf.get("data_types", "kaldi_ark,text")
- dataset = AudioDataset(scp_lists, data_names, data_types, frontend_conf=frontend_conf, shuffle=shuffle, mode=mode)
+
+ pre_hwfile = conf.get("pre_hwlist", None)
+ pre_prob = conf.get("pre_prob", 0)
+
+ hw_config = {"sample_rate": conf.get("sample_rate", 0.6),
+ "double_rate": conf.get("double_rate", 0.1),
+ "hotword_min_length": conf.get("hotword_min_length", 2),
+ "hotword_max_length": conf.get("hotword_max_length", 8)}
+
+
+ if pre_hwfile is not None:
+ pre_hwlist = []
+ with open(pre_hwfile, 'r') as fin:
+ for line in fin.readlines():
+ pre_hwlist.append(line.strip())
+ else:
+ pre_hwlist = None
+ # logging.warning("Previous hwlist: {}".format(pre_hwlist))
+ dataset = AudioDataset(scp_lists,
+ data_names,
+ data_types,
+ frontend_conf=frontend_conf,
+ shuffle=shuffle,
+ mode=mode,
+ pre_hwlist=pre_hwlist,
+ pre_prob=pre_prob)
filter_conf = conf.get('filter_conf', {})
filter_fn = partial(filter, **filter_conf)
dataset = FilterIterDataPipe(dataset, fn=filter_fn)
if "text" in data_names:
- vocab = {'vocab': dict, 'seg_dict': seg_dict, 'punc_dict': punc_dict, 'bpe_tokenizer': bpe_tokenizer}
+ vocab = {'vocab': dict, 'seg_dict': seg_dict, 'punc_dict': punc_dict, 'bpe_tokenizer': bpe_tokenizer, 'hw_config': hw_config}
tokenize_fn = partial(tokenize, **vocab)
dataset = MapperIterDataPipe(dataset, fn=tokenize_fn)
diff --git a/funasr/datasets/large_datasets/utils/hotword_utils.py b/funasr/datasets/large_datasets/utils/hotword_utils.py
new file mode 100644
index 0000000..fccfea6
--- /dev/null
+++ b/funasr/datasets/large_datasets/utils/hotword_utils.py
@@ -0,0 +1,32 @@
+import random
+
+def sample_hotword(length,
+ hotword_min_length,
+ hotword_max_length,
+ sample_rate,
+ double_rate,
+ pre_prob,
+ pre_index=None):
+ if length < hotword_min_length:
+ return [-1]
+ if random.random() < sample_rate:
+ if pre_prob > 0 and random.random() < pre_prob and pre_index is not None:
+ return pre_index
+ if length == hotword_min_length:
+ return [0, length-1]
+ elif random.random() < double_rate and length > hotword_max_length + hotword_min_length + 2:
+ # sample two hotwords in a sentence
+ _max_hw_length = min(hotword_max_length, length // 2)
+ # first hotword
+ start1 = random.randint(0, length // 3)
+ end1 = random.randint(start1 + hotword_min_length - 1, start1 + _max_hw_length - 1)
+ # second hotword
+ start2 = random.randint(end1 + 1, length - hotword_min_length)
+ end2 = random.randint(min(length-1, start2+hotword_min_length-1), min(length-1, start2+hotword_max_length-1))
+ return [start1, end1, start2, end2]
+ else: # single hotword
+ start = random.randint(0, length - hotword_min_length)
+ end = random.randint(min(length-1, start+hotword_min_length-1), min(length-1, start+hotword_max_length-1))
+ return [start, end]
+ else:
+ return [-1]
\ No newline at end of file
diff --git a/funasr/datasets/large_datasets/utils/padding.py b/funasr/datasets/large_datasets/utils/padding.py
index e0feac6..fdca63d 100644
--- a/funasr/datasets/large_datasets/utils/padding.py
+++ b/funasr/datasets/large_datasets/utils/padding.py
@@ -31,4 +31,47 @@
batch[data_name] = tensor_pad
batch[data_name + "_lengths"] = tensor_lengths
+ # DHA, EAHC NOT INCLUDED
+ if "hotword_indxs" in batch:
+ # if hotword indxs in batch
+ # use it to slice hotwords out
+ hotword_list = []
+ hotword_lengths = []
+ text = batch['text']
+ text_lengths = batch['text_lengths']
+ hotword_indxs = batch['hotword_indxs']
+ num_hw = sum([int(i) for i in batch['hotword_indxs_lengths'] if i != 1]) // 2
+ B, t1 = text.shape
+ t1 += 1 # TODO: as parameter which is same as predictor_bias
+ ideal_attn = torch.zeros(B, t1, num_hw+1)
+ nth_hw = 0
+ for b, (hotword_indx, one_text, length) in enumerate(zip(hotword_indxs, text, text_lengths)):
+ ideal_attn[b][:,-1] = 1
+ if hotword_indx[0] != -1:
+ start, end = int(hotword_indx[0]), int(hotword_indx[1])
+ hotword = one_text[start: end+1]
+ hotword_list.append(hotword)
+ hotword_lengths.append(end-start+1)
+ ideal_attn[b][start:end+1, nth_hw] = 1
+ ideal_attn[b][start:end+1, -1] = 0
+ nth_hw += 1
+ if len(hotword_indx) == 4 and hotword_indx[2] != -1:
+ # the second hotword if exist
+ start, end = int(hotword_indx[2]), int(hotword_indx[3])
+ hotword_list.append(one_text[start: end+1])
+ hotword_lengths.append(end-start+1)
+ ideal_attn[b][start:end+1, nth_hw-1] = 1
+ ideal_attn[b][start:end+1, -1] = 0
+ nth_hw += 1
+ hotword_list.append(torch.tensor([1]))
+ hotword_lengths.append(1)
+ hotword_pad = pad_sequence(hotword_list,
+ batch_first=True,
+ padding_value=0)
+ batch["hotword_pad"] = hotword_pad
+ batch["hotword_lengths"] = torch.tensor(hotword_lengths, dtype=torch.int32)
+ batch['ideal_attn'] = ideal_attn
+ del batch['hotword_indxs']
+ del batch['hotword_indxs_lengths']
+
return keys, batch
diff --git a/funasr/datasets/large_datasets/utils/tokenize.py b/funasr/datasets/large_datasets/utils/tokenize.py
index 0d2fd84..09ece76 100644
--- a/funasr/datasets/large_datasets/utils/tokenize.py
+++ b/funasr/datasets/large_datasets/utils/tokenize.py
@@ -1,6 +1,7 @@
#!/usr/bin/env python
import re
import numpy as np
+from funasr.datasets.large_datasets.utils.hotword_utils import sample_hotword
def forward_segment(text, seg_dict):
word_list = []
@@ -38,7 +39,8 @@
vocab=None,
seg_dict=None,
punc_dict=None,
- bpe_tokenizer=None):
+ bpe_tokenizer=None,
+ hw_config=None):
assert "text" in data
assert isinstance(vocab, dict)
text = data["text"]
@@ -53,6 +55,9 @@
text = seg_tokenize(text, seg_dict)
length = len(text)
+ if 'hw_tag' in data:
+ hotword_indxs = sample_hotword(length, **hw_config)
+ data[hotword_indxs] = hotword_indxs
for i in range(length):
x = text[i]
if i == length-1 and "punc" in data and x.startswith("vad:"):
diff --git a/funasr/models/e2e_asr_contextual_paraformer.py b/funasr/models/e2e_asr_contextual_paraformer.py
new file mode 100644
index 0000000..cafb653
--- /dev/null
+++ b/funasr/models/e2e_asr_contextual_paraformer.py
@@ -0,0 +1,408 @@
+from json import decoder
+import logging
+from contextlib import contextmanager
+from distutils.version import LooseVersion
+from typing import Dict
+from typing import List
+from typing import Optional
+from typing import Tuple
+from typing import Union
+import random
+from unicodedata import bidirectional
+import numpy as np
+
+import torch
+from typeguard import check_argument_types
+
+from funasr.layers.abs_normalize import AbsNormalize
+from funasr.losses.label_smoothing_loss import (
+ LabelSmoothingLoss, # noqa: H301
+)
+from funasr.models.ctc import CTC
+from funasr.models.decoder.abs_decoder import AbsDecoder
+from funasr.models.e2e_asr_common import ErrorCalculator
+from funasr.models.encoder.abs_encoder import AbsEncoder
+from funasr.models.frontend.abs_frontend import AbsFrontend
+from funasr.models.postencoder.abs_postencoder import AbsPostEncoder
+from funasr.models.predictor.cif import mae_loss
+from funasr.models.preencoder.abs_preencoder import AbsPreEncoder
+from funasr.models.specaug.abs_specaug import AbsSpecAug
+from funasr.modules.add_sos_eos import add_sos_eos
+from funasr.modules.nets_utils import make_pad_mask, pad_list
+from funasr.modules.nets_utils import th_accuracy
+from funasr.torch_utils.device_funcs import force_gatherable
+from funasr.train.abs_espnet_model import AbsESPnetModel
+from funasr.models.predictor.cif import CifPredictorV3
+from funasr.modules.streaming_utils import utils as myutils
+from funasr.models.e2e_asr_paraformer import Paraformer
+from funasr.modules.layer_norm import LayerNorm
+
+
+if LooseVersion(torch.__version__) >= LooseVersion("1.6.0"):
+ from torch.cuda.amp import autocast
+else:
+ # Nothing to do if torch<1.6.0
+ @contextmanager
+ def autocast(enabled=True):
+ yield
+
+
+class AdvancedContextualParaformer(Paraformer):
+ def __init__(
+ self,
+ vocab_size: int,
+ token_list: Union[Tuple[str, ...], List[str]],
+ frontend: Optional[AbsFrontend],
+ specaug: Optional[AbsSpecAug],
+ normalize: Optional[AbsNormalize],
+ preencoder: Optional[AbsPreEncoder],
+ encoder: AbsEncoder,
+ postencoder: Optional[AbsPostEncoder],
+ decoder: AbsDecoder,
+ ctc: CTC,
+ ctc_weight: float = 0.5,
+ interctc_weight: float = 0.0,
+ ignore_id: int = -1,
+ blank_id: int = 0,
+ sos: int = 1,
+ eos: int = 2,
+ lsm_weight: float = 0.0,
+ length_normalized_loss: bool = False,
+ report_cer: bool = True,
+ report_wer: bool = True,
+ sym_space: str = "<space>",
+ sym_blank: str = "<blank>",
+ extract_feats_in_collect_stats: bool = True,
+ predictor = None,
+ predictor_weight: float = 0.0,
+ predictor_bias: int = 0,
+ sampling_ratio: float = 0.2,
+ target_buffer_length: int = -1,
+ inner_dim: int = 256,
+ bias_encoder_type: str = 'lstm',
+ use_decoder_embedding: bool = True,
+ crit_attn_weight: float = 0.0,
+ crit_attn_smooth: float = 0.0,
+ bias_encoder_dropout_rate: float = 0.0,
+ ):
+ assert check_argument_types()
+ assert 0.0 <= ctc_weight <= 1.0, ctc_weight
+ assert 0.0 <= interctc_weight < 1.0, interctc_weight
+
+ super().__init__(
+ vocab_size=vocab_size,
+ token_list=token_list,
+ frontend=frontend,
+ specaug=specaug,
+ normalize=normalize,
+ preencoder=preencoder,
+ encoder=encoder,
+ postencoder=postencoder,
+ decoder=decoder,
+ ctc=ctc,
+ ctc_weight=ctc_weight,
+ interctc_weight=interctc_weight,
+ ignore_id=ignore_id,
+ blank_id=blank_id,
+ sos=sos,
+ eos=eos,
+ lsm_weight=lsm_weight,
+ length_normalized_loss=length_normalized_loss,
+ report_cer=report_cer,
+ report_wer=report_wer,
+ sym_space=sym_space,
+ sym_blank=sym_blank,
+ extract_feats_in_collect_stats=extract_feats_in_collect_stats,
+ predictor=predictor,
+ predictor_weight=predictor_weight,
+ predictor_bias=predictor_bias,
+ sampling_ratio=sampling_ratio,
+ )
+
+ if bias_encoder_type == 'lstm':
+ logging.warning("enable bias encoder sampling and contextual training")
+ self.bias_encoder = torch.nn.LSTM(inner_dim, inner_dim, 1, batch_first=True, dropout=bias_encoder_dropout_rate)
+ self.bias_embed = torch.nn.Embedding(vocab_size, inner_dim)
+ elif bias_encoder_type == 'mean':
+ logging.warning("enable bias encoder sampling and contextual training")
+ self.bias_embed = torch.nn.Embedding(vocab_size, inner_dim)
+ else:
+ logging.error("Unsupport bias encoder type: {}".format(bias_encoder_type))
+
+ self.target_buffer_length = target_buffer_length
+ if self.target_buffer_length > 0:
+ self.hotword_buffer = None
+ self.length_record = []
+ self.current_buffer_length = 0
+ self.use_decoder_embedding = use_decoder_embedding
+ self.crit_attn_weight = crit_attn_weight
+ if self.crit_attn_weight > 0:
+ self.attn_loss = torch.nn.L1Loss()
+ self.crit_attn_smooth = crit_attn_smooth
+
+ def forward(
+ self,
+ speech: torch.Tensor,
+ speech_lengths: torch.Tensor,
+ text: torch.Tensor,
+ text_lengths: torch.Tensor,
+ hotword_pad: torch.Tensor,
+ hotword_lengths: torch.Tensor,
+ ideal_attn: torch.Tensor,
+ ) -> Tuple[torch.Tensor, Dict[str, torch.Tensor], torch.Tensor]:
+ """Frontend + Encoder + Decoder + Calc loss
+
+ Args:
+ speech: (Batch, Length, ...)
+ speech_lengths: (Batch, )
+ text: (Batch, Length)
+ text_lengths: (Batch,)
+ """
+ assert text_lengths.dim() == 1, text_lengths.shape
+ # Check that batch_size is unified
+ assert (
+ speech.shape[0]
+ == speech_lengths.shape[0]
+ == text.shape[0]
+ == text_lengths.shape[0]
+ ), (speech.shape, speech_lengths.shape, text.shape, text_lengths.shape)
+ batch_size = speech.shape[0]
+ self.step_cur += 1
+ # for data-parallel
+ text = text[:, : text_lengths.max()]
+ speech = speech[:, :speech_lengths.max()]
+
+ # 1. Encoder
+ encoder_out, encoder_out_lens = self.encode(speech, speech_lengths)
+ intermediate_outs = None
+ if isinstance(encoder_out, tuple):
+ intermediate_outs = encoder_out[1]
+ encoder_out = encoder_out[0]
+
+ loss_att, acc_att, cer_att, wer_att = None, None, None, None
+ loss_ctc, cer_ctc = None, None
+ loss_pre = None
+ loss_ideal = None
+
+ stats = dict()
+
+ # 1. CTC branch
+ if self.ctc_weight != 0.0:
+ loss_ctc, cer_ctc = self._calc_ctc_loss(
+ encoder_out, encoder_out_lens, 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
+
+ # Intermediate CTC (optional)
+ loss_interctc = 0.0
+ if self.interctc_weight != 0.0 and intermediate_outs is not None:
+ for layer_idx, intermediate_out in intermediate_outs:
+ # we assume intermediate_out has the same length & padding
+ # as those of encoder_out
+ loss_ic, cer_ic = self._calc_ctc_loss(
+ intermediate_out, encoder_out_lens, text, text_lengths
+ )
+ loss_interctc = loss_interctc + loss_ic
+
+ # Collect Intermedaite CTC stats
+ stats["loss_interctc_layer{}".format(layer_idx)] = (
+ loss_ic.detach() if loss_ic is not None else None
+ )
+ stats["cer_interctc_layer{}".format(layer_idx)] = cer_ic
+
+ loss_interctc = loss_interctc / len(intermediate_outs)
+
+ # calculate whole encoder loss
+ loss_ctc = (1 - self.interctc_weight) * loss_ctc + self.interctc_weight * loss_interctc
+
+ # 2b. Attention decoder branch
+ if self.ctc_weight != 1.0:
+ loss_att, acc_att, cer_att, wer_att, loss_pre, loss_ideal = self._calc_att_clas_loss(
+ encoder_out, encoder_out_lens, text, text_lengths, hotword_pad, hotword_lengths, ideal_attn
+ )
+
+ # 3. CTC-Att loss definition
+ if self.ctc_weight == 0.0:
+ loss = loss_att + loss_pre * self.predictor_weight
+ elif self.ctc_weight == 1.0:
+ loss = loss_ctc
+ else:
+ loss = self.ctc_weight * loss_ctc + (1 - self.ctc_weight) * loss_att + loss_pre * self.predictor_weight
+
+ if loss_ideal is not None:
+ loss = loss + loss_ideal * self.crit_attn_weight
+ stats["loss_ideal"] = loss_ideal.detach().cpu()
+
+ # Collect Attn branch stats
+ stats["loss_att"] = loss_att.detach() if 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
+ loss, stats, weight = force_gatherable((loss, stats, batch_size), loss.device)
+ return loss, stats, weight
+
+ def _calc_att_clas_loss(
+ self,
+ encoder_out: torch.Tensor,
+ encoder_out_lens: torch.Tensor,
+ ys_pad: torch.Tensor,
+ ys_pad_lens: torch.Tensor,
+ hotword_pad: torch.Tensor,
+ hotword_lengths: torch.Tensor,
+ ideal_attn: 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
+ pre_acoustic_embeds, pre_token_length, _, _ = self.predictor(encoder_out, ys_pad, encoder_out_mask,
+ ignore_id=self.ignore_id)
+
+ # -1. bias encoder
+ if self.use_decoder_embedding:
+ hw_embed = self.decoder.embed(hotword_pad)
+ else:
+ hw_embed = self.bias_embed(hotword_pad)
+ hw_embed, (_, _) = self.bias_encoder(hw_embed)
+ _ind = np.arange(0, hotword_pad.shape[0]).tolist()
+ selected = hw_embed[_ind, [i-1 for i in hotword_lengths.detach().cpu().tolist()]]
+ contextual_info = selected.squeeze(0).repeat(ys_pad.shape[0], 1, 1).to(ys_pad.device)
+
+ # 0. sampler
+ decoder_out_1st = None
+ if self.sampling_ratio > 0.0:
+ if self.step_cur < 2:
+ logging.info("enable sampler in paraformer, sampling_ratio: {}".format(self.sampling_ratio))
+ sematic_embeds, decoder_out_1st = self.sampler(encoder_out, encoder_out_lens, ys_pad, ys_pad_lens,
+ pre_acoustic_embeds, contextual_info)
+ 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, contextual_info=contextual_info, ret_attn=(ideal_attn is not None)
+ )
+ decoder_out, _, attn = decoder_outs[0], decoder_outs[1], decoder_outs[2]
+ if self.crit_attn_weight > 0 and attn.shape[-1] > 1:
+ ideal_attn = ideal_attn + self.crit_attn_smooth / (self.crit_attn_smooth + 1.0)
+ attn_non_blank = attn[:,:,:,:-1]
+ ideal_attn_non_blank = ideal_attn[:,:,:-1]
+ loss_ideal = self.attn_loss(attn_non_blank.max(1)[0], ideal_attn_non_blank.to(attn.device))
+ else:
+ loss_ideal = None
+
+ 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, loss_ideal
+
+ def sampler(self, encoder_out, encoder_out_lens, ys_pad, ys_pad_lens, pre_acoustic_embeds, contextual_info):
+
+ tgt_mask = (~make_pad_mask(ys_pad_lens, maxlen=ys_pad_lens.max())[:, :, None]).to(ys_pad.device)
+ ys_pad = ys_pad * tgt_mask[:, :, 0]
+ if self.share_embedding:
+ ys_pad_embed = self.decoder.output_layer.weight[ys_pad]
+ else:
+ ys_pad_embed = self.decoder.embed(ys_pad)
+ with torch.no_grad():
+ decoder_outs = self.decoder(
+ encoder_out, encoder_out_lens, pre_acoustic_embeds, ys_pad_lens, contextual_info=contextual_info
+ )
+ 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].to(pre_acoustic_embeds.device), 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 cal_decoder_with_predictor_with_hwlist_advanced(self, encoder_out, encoder_out_lens, sematic_embeds, ys_pad_lens, hw_list=None):
+ if hw_list is None:
+ hw_list = [torch.Tensor([1]).long().to(encoder_out.device)] # empty hotword list
+ hw_list_pad = pad_list(hw_list, 0)
+ if self.use_decoder_embedding:
+ hw_embed = self.decoder.embed(hw_list_pad)
+ else:
+ hw_embed = self.bias_embed(hw_list_pad)
+ hw_embed, (h_n, _) = self.bias_encoder(hw_embed)
+ else:
+ # hw_list = hw_list[1:] + [hw_list[0]] # reorder
+ 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)
+ if self.use_decoder_embedding:
+ hw_embed = self.decoder.embed(hw_list_pad)
+ else:
+ hw_embed = self.bias_embed(hw_list_pad)
+ hw_embed = torch.nn.utils.rnn.pack_padded_sequence(hw_embed, hw_lengths, batch_first=True,
+ enforce_sorted=False)
+ _, (h_n, _) = self.bias_encoder(hw_embed)
+ # hw_embed, _ = torch.nn.utils.rnn.pad_packed_sequence(hw_embed, batch_first=True)
+ if h_n.shape[1] > 2000: # large hotword list
+ _h_n = self.pick_hwlist_group(h_n.squeeze(0), encoder_out, encoder_out_lens, sematic_embeds, ys_pad_lens)
+ if _h_n is not None:
+ h_n = _h_n
+ hw_embed = h_n.repeat(encoder_out.shape[0], 1, 1)
+ # import pdb; pdb.set_trace()
+
+ decoder_outs = self.decoder(
+ encoder_out, encoder_out_lens, sematic_embeds, ys_pad_lens, contextual_info=hw_embed
+ )
+ decoder_out = decoder_outs[0]
+ decoder_out = torch.log_softmax(decoder_out, dim=-1)
+ return decoder_out, ys_pad_lens
+
+ def pick_hwlist_group(self, hw_embed, encoder_out, encoder_out_lens, sematic_embeds, ys_pad_lens):
+ max_attn_score = 0.0
+ # max_attn_index = 0
+ argmax_g = None
+ non_blank = hw_embed[-1]
+ hw_embed_groups = hw_embed[:-1].split(2000)
+ for i, g in enumerate(hw_embed_groups):
+ g = torch.cat([g, non_blank.unsqueeze(0)], dim=0)
+ _ = self.decoder(
+ encoder_out, encoder_out_lens, sematic_embeds, ys_pad_lens, contextual_info=g.unsqueeze(0)
+ )
+ attn = self.decoder.bias_decoder.src_attn.attn[0]
+ _max_attn_score = attn.max(0)[0][:,:-1].max()
+ if _max_attn_score > max_attn_score:
+ max_attn_score = _max_attn_score
+ # max_attn_index = i
+ argmax_g = g
+ # import pdb; pdb.set_trace()
+ return argmax_g
\ No newline at end of file
diff --git a/funasr/tasks/asr.py b/funasr/tasks/asr.py
index d52c9c3..9e33c11 100644
--- a/funasr/tasks/asr.py
+++ b/funasr/tasks/asr.py
@@ -42,6 +42,7 @@
from funasr.models.joint_net.joint_network import JointNetwork
from funasr.models.e2e_asr import ESPnetASRModel
from funasr.models.e2e_asr_paraformer import Paraformer, ParaformerOnline, ParaformerBert, BiCifParaformer, ContextualParaformer
+from funasr.models.e2e_asr_contextual_paraformer import AdvancedContextualParaformer
from funasr.models.e2e_tp import TimestampPredictor
from funasr.models.e2e_asr_mfcca import MFCCA
from funasr.models.e2e_uni_asr import UniASR
@@ -128,6 +129,7 @@
paraformer_bert=ParaformerBert,
bicif_paraformer=BiCifParaformer,
contextual_paraformer=ContextualParaformer,
+ acontextual_paraformer=AdvancedContextualParaformer,
mfcca=MFCCA,
timestamp_prediction=TimestampPredictor,
),
--
Gitblit v1.9.1