From 39efd2204c84b086dceac3dbb257425fc640a3ff Mon Sep 17 00:00:00 2001
From: Lizerui9926 <110582652+Lizerui9926@users.noreply.github.com>
Date: 星期一, 17 四月 2023 11:05:01 +0800
Subject: [PATCH] Merge pull request #364 from alibaba-damo-academy/dev_sx
---
funasr/models/e2e_asr_paraformer.py | 147 +++++++++++++++++++++++++++++++++++++++++++-----
1 files changed, 131 insertions(+), 16 deletions(-)
diff --git a/funasr/models/e2e_asr_paraformer.py b/funasr/models/e2e_asr_paraformer.py
index 02f60af..fcef342 100644
--- a/funasr/models/e2e_asr_paraformer.py
+++ b/funasr/models/e2e_asr_paraformer.py
@@ -44,7 +44,7 @@
class Paraformer(AbsESPnetModel):
"""
- Author: Speech Lab, Alibaba Group, China
+ 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
"""
@@ -370,19 +370,10 @@
encoder_out, encoder_out_lens
)
- assert encoder_out.size(0) == speech.size(0), (
- encoder_out.size(),
- speech.size(0),
- )
- assert encoder_out.size(1) <= encoder_out_lens.max(), (
- encoder_out.size(),
- encoder_out_lens.max(),
- )
-
if intermediate_outs is not None:
return (encoder_out, intermediate_outs), encoder_out_lens
- return encoder_out, encoder_out_lens
+ return encoder_out, torch.tensor([encoder_out.size(1)])
def calc_predictor(self, encoder_out, encoder_out_lens):
@@ -621,7 +612,7 @@
class ParaformerBert(Paraformer):
"""
- Author: Speech Lab, Alibaba Group, China
+ Author: Speech Lab of DAMO Academy, Alibaba Group
Paraformer2: advanced paraformer with LFMMI and bert for non-autoregressive end-to-end speech recognition
"""
@@ -986,6 +977,59 @@
loss_pre2 = self.criterion_pre(ys_pad_lens.type_as(pre_token_length2), pre_token_length2)
return loss_pre2
+
+ def _calc_att_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
+ pre_acoustic_embeds, pre_token_length, _, pre_peak_index, _ = self.predictor(encoder_out, ys_pad, encoder_out_mask,
+ ignore_id=self.ignore_id)
+
+ # 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)
+ 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
+ )
+ 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
def calc_predictor(self, encoder_out, encoder_out_lens):
@@ -1034,16 +1078,76 @@
# 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
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 = self._calc_att_loss(
+ encoder_out, encoder_out_lens, text, text_lengths
+ )
loss_pre2 = self._calc_pre2_loss(
encoder_out, encoder_out_lens, text, text_lengths
)
- loss = loss_pre2
+ # 3. CTC-Att loss definition
+ if self.ctc_weight == 0.0:
+ loss = loss_att + loss_pre * self.predictor_weight + loss_pre2 * self.predictor_weight * 0.5
+ 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 + loss_pre2 * self.predictor_weight * 0.5
+ # 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_pre2"] = loss_pre2.detach().cpu()
+
stats["loss"] = torch.clone(loss.detach())
# force_gatherable: to-device and to-tensor if scalar for DataParallel
@@ -1094,6 +1198,7 @@
inner_dim: int = 256,
bias_encoder_type: str = 'lstm',
label_bracket: bool = False,
+ use_decoder_embedding: bool = False,
):
assert check_argument_types()
assert 0.0 <= ctc_weight <= 1.0, ctc_weight
@@ -1147,6 +1252,7 @@
self.hotword_buffer = None
self.length_record = []
self.current_buffer_length = 0
+ self.use_decoder_embedding = use_decoder_embedding
def forward(
self,
@@ -1288,7 +1394,10 @@
hw_list.append(hw_tokens)
# padding
hw_list_pad = pad_list(hw_list, 0)
- hw_embed = self.decoder.embed(hw_list_pad)
+ if self.use_decoder_embedding:
+ hw_embed = self.decoder.embed(hw_list_pad)
+ else:
+ hw_embed = self.bias_embed(hw_list_pad)
hw_embed, (_, _) = self.bias_encoder(hw_embed)
_ind = np.arange(0, len(hw_list)).tolist()
# update self.hotword_buffer, throw a part if oversize
@@ -1404,13 +1513,19 @@
# default hotword list
hw_list = [torch.Tensor([self.sos]).long().to(encoder_out.device)] # empty hotword list
hw_list_pad = pad_list(hw_list, 0)
- hw_embed = self.bias_embed(hw_list_pad)
+ if self.use_decoder_embedding:
+ hw_embed = self.decoder.embed(hw_list_pad)
+ else:
+ hw_embed = self.bias_embed(hw_list_pad)
_, (h_n, _) = self.bias_encoder(hw_embed)
contextual_info = h_n.squeeze(0).repeat(encoder_out.shape[0], 1, 1)
else:
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)
- hw_embed = self.bias_embed(hw_list_pad)
+ 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)
--
Gitblit v1.9.1