From 2cfe010d7b0f17877a271cc401e2c2f8f8d4c42c Mon Sep 17 00:00:00 2001
From: speech_asr <wangjiaming.wjm@alibaba-inc.com>
Date: 星期三, 15 三月 2023 14:42:32 +0800
Subject: [PATCH] update
---
funasr/models/e2e_diar_eend_ola.py | 28 ++++++++++++++++++----------
1 files changed, 18 insertions(+), 10 deletions(-)
diff --git a/funasr/models/e2e_diar_eend_ola.py b/funasr/models/e2e_diar_eend_ola.py
index 2960b23..f3e34bc 100644
--- a/funasr/models/e2e_diar_eend_ola.py
+++ b/funasr/models/e2e_diar_eend_ola.py
@@ -11,7 +11,8 @@
import torch.nn as nn
from typeguard import check_argument_types
-from funasr.modules.eend_ola.encoder import TransformerEncoder
+from funasr.models.frontend.wav_frontend import WavFrontendMel23
+from funasr.modules.eend_ola.encoder import EENDOLATransformerEncoder
from funasr.modules.eend_ola.encoder_decoder_attractor import EncoderDecoderAttractor
from funasr.modules.eend_ola.utils.power import generate_mapping_dict
from funasr.torch_utils.device_funcs import force_gatherable
@@ -34,12 +35,13 @@
class DiarEENDOLAModel(AbsESPnetModel):
- """CTC-attention hybrid Encoder-Decoder model"""
+ """EEND-OLA diarization model"""
def __init__(
self,
- encoder: TransformerEncoder,
- eda: EncoderDecoderAttractor,
+ frontend: WavFrontendMel23,
+ encoder: EENDOLATransformerEncoder,
+ encoder_decoder_attractor: EncoderDecoderAttractor,
n_units: int = 256,
max_n_speaker: int = 8,
attractor_loss_weight: float = 1.0,
@@ -49,15 +51,16 @@
assert check_argument_types()
super().__init__()
- self.encoder = encoder
- self.eda = eda
+ self.frontend = frontend
+ self.enc = encoder
+ self.eda = encoder_decoder_attractor
self.attractor_loss_weight = attractor_loss_weight
self.max_n_speaker = max_n_speaker
if mapping_dict is None:
mapping_dict = generate_mapping_dict(max_speaker_num=self.max_n_speaker)
self.mapping_dict = mapping_dict
# PostNet
- self.PostNet = nn.LSTM(self.max_n_speaker, n_units, 1, batch_first=True)
+ self.postnet = nn.LSTM(self.max_n_speaker, n_units, 1, batch_first=True)
self.output_layer = nn.Linear(n_units, mapping_dict['oov'] + 1)
def forward_encoder(self, xs, ilens):
@@ -65,7 +68,7 @@
pad_shape = xs.shape
xs_mask = [torch.ones(ilen).to(xs.device) for ilen in ilens]
xs_mask = torch.nn.utils.rnn.pad_sequence(xs_mask, batch_first=True, padding_value=0).unsqueeze(-2)
- emb = self.encoder(xs, xs_mask)
+ emb = self.enc(xs, xs_mask)
emb = torch.split(emb.view(pad_shape[0], pad_shape[1], -1), 1, dim=0)
emb = [e[0][:ilen] for e, ilen in zip(emb, ilens)]
return emb
@@ -74,7 +77,7 @@
maxlen = torch.max(ilens).to(torch.int).item()
logits = nn.utils.rnn.pad_sequence(logits, batch_first=True, padding_value=-1)
logits = nn.utils.rnn.pack_padded_sequence(logits, ilens, batch_first=True, enforce_sorted=False)
- outputs, (_, _) = self.PostNet(logits)
+ outputs, (_, _) = self.postnet(logits)
outputs = nn.utils.rnn.pad_packed_sequence(outputs, batch_first=True, padding_value=-1, total_length=maxlen)[0]
outputs = [output[:ilens[i].to(torch.int).item()] for i, output in enumerate(outputs)]
outputs = [self.output_layer(output) for output in outputs]
@@ -109,7 +112,7 @@
text = text[:, : text_lengths.max()]
# 1. Encoder
- encoder_out, encoder_out_lens = self.encode(speech, speech_lengths)
+ encoder_out, encoder_out_lens = self.enc(speech, speech_lengths)
intermediate_outs = None
if isinstance(encoder_out, tuple):
intermediate_outs = encoder_out[1]
@@ -187,6 +190,8 @@
shuffle: bool = True,
threshold: float = 0.5,
**kwargs):
+ if self.frontend is not None:
+ speech = self.frontend(speech)
speech = [s[:s_len] for s, s_len in zip(speech, speech_lengths)]
emb = self.forward_encoder(speech, speech_lengths)
if shuffle:
@@ -235,3 +240,6 @@
torch.float32)
decisions = decisions[:, :n_speaker]
return decisions
+
+ def collect_feats(self, **batch: torch.Tensor) -> Dict[str, torch.Tensor]:
+ pass
\ No newline at end of file
--
Gitblit v1.9.1