From 701022837aa295869a6f39be0db42d4e18fdf160 Mon Sep 17 00:00:00 2001
From: 嘉渊 <wangjiaming.wjm@alibaba-inc.com>
Date: 星期三, 17 五月 2023 19:53:00 +0800
Subject: [PATCH] update repo
---
egs/aishell/paraformer/conf/train_asr_paraformer_conformer_12e_6d_2048_256.yaml | 1 +
funasr/models/e2e_asr_paraformer.py | 52 +++++++++++++++++++++++++++++++++++++++++++++++-----
2 files changed, 48 insertions(+), 5 deletions(-)
diff --git a/egs/aishell/paraformer/conf/train_asr_paraformer_conformer_12e_6d_2048_256.yaml b/egs/aishell/paraformer/conf/train_asr_paraformer_conformer_12e_6d_2048_256.yaml
index 6a14b7f..9dd3fb3 100644
--- a/egs/aishell/paraformer/conf/train_asr_paraformer_conformer_12e_6d_2048_256.yaml
+++ b/egs/aishell/paraformer/conf/train_asr_paraformer_conformer_12e_6d_2048_256.yaml
@@ -47,6 +47,7 @@
length_normalized_loss: false
predictor_weight: 1.0
sampling_ratio: 0.4
+ use_1st_decoder_loss: true
# optimization related
accum_grad: 1
diff --git a/funasr/models/e2e_asr_paraformer.py b/funasr/models/e2e_asr_paraformer.py
index 00e08b1..060442c 100644
--- a/funasr/models/e2e_asr_paraformer.py
+++ b/funasr/models/e2e_asr_paraformer.py
@@ -78,6 +78,7 @@
share_embedding: bool = False,
preencoder: Optional[AbsPreEncoder] = None,
postencoder: Optional[AbsPostEncoder] = None,
+ use_1st_decoder_loss: bool = False,
):
assert check_argument_types()
assert 0.0 <= ctc_weight <= 1.0, ctc_weight
@@ -144,6 +145,8 @@
if self.share_embedding:
self.decoder.embed = None
+ self.use_1st_decoder_loss = use_1st_decoder_loss
+
def forward(
self,
speech: torch.Tensor,
@@ -179,7 +182,7 @@
intermediate_outs = encoder_out[1]
encoder_out = encoder_out[0]
- loss_att, acc_att, cer_att, wer_att = None, None, None, None
+ loss_att, pre_loss_att, acc_att, cer_att, wer_att = None, None, None, None, None
loss_ctc, cer_ctc = None, None
loss_pre = None
stats = dict()
@@ -220,7 +223,7 @@
# 2b. Attention decoder branch
if self.ctc_weight != 1.0:
- loss_att, acc_att, cer_att, wer_att, loss_pre = self._calc_att_loss(
+ loss_att, acc_att, cer_att, wer_att, loss_pre, pre_loss_att = self._calc_att_loss(
encoder_out, encoder_out_lens, text, text_lengths
)
@@ -231,6 +234,9 @@
loss = loss_ctc
else:
loss = self.ctc_weight * loss_ctc + (1 - self.ctc_weight) * loss_att + loss_pre * self.predictor_weight
+
+ if self.use_1st_decoder_loss and pre_loss_att is not None:
+ loss = loss + pre_loss_att
# Collect Attn branch stats
stats["loss_att"] = loss_att.detach() if loss_att is not None else None
@@ -456,11 +462,16 @@
# 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))
- sematic_embeds, decoder_out_1st = self.sampler(encoder_out, encoder_out_lens, ys_pad, ys_pad_lens,
- pre_acoustic_embeds)
+ 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)
+ else:
+ 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))
@@ -490,7 +501,7 @@
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
+ 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):
@@ -523,6 +534,37 @@
input_mask_expand_dim, 0)
return sematic_embeds * tgt_mask, decoder_out * tgt_mask
+ def sampler_with_grad(self, encoder_out, encoder_out_lens, ys_pad, ys_pad_lens, pre_acoustic_embeds):
+ 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)
+ decoder_outs = self.decoder(
+ encoder_out, encoder_out_lens, pre_acoustic_embeds, ys_pad_lens
+ )
+ pre_loss_att = self.criterion_att(decoder_outs[0], ys_pad)
+ 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, pre_loss_att
+
def _calc_ctc_loss(
self,
encoder_out: torch.Tensor,
--
Gitblit v1.9.1