# Copyright 2022 Jiatong Shi (Carnegie Mellon University)
# Apache 2.0 (http://www.apache.org/licenses/LICENSE-2.0)
from contextlib import contextmanager
from typing import Dict, Optional, Tuple
import torch
from packaging.version import parse as V
from typeguard import typechecked
from espnet2.asr.encoder.abs_encoder import AbsEncoder
from espnet2.asr.frontend.abs_frontend import AbsFrontend
from espnet2.asr.preencoder.abs_preencoder import AbsPreEncoder
from espnet2.asr.specaug.abs_specaug import AbsSpecAug
from espnet2.asvspoof.decoder.abs_decoder import AbsDecoder
from espnet2.asvspoof.loss.abs_loss import AbsASVSpoofLoss
from espnet2.layers.abs_normalize import AbsNormalize
from espnet2.torch_utils.device_funcs import force_gatherable
from espnet2.train.abs_espnet_model import AbsESPnetModel
if V(torch.__version__) >= V("1.6.0"):
from torch.cuda.amp import autocast
else:
# Nothing to do if torch<1.6.0
@contextmanager
def autocast(enabled=True):
yield
[docs]class ESPnetASVSpoofModel(AbsESPnetModel):
"""ASV Spoofing model
A simple ASV Spoofing model
"""
@typechecked
def __init__(
self,
frontend: Optional[AbsFrontend],
specaug: Optional[AbsSpecAug],
normalize: Optional[AbsNormalize],
encoder: AbsEncoder,
preencoder: Optional[AbsPreEncoder],
decoder: AbsDecoder,
losses: Dict[str, AbsASVSpoofLoss],
):
super().__init__()
self.preencoder = preencoder
self.encoder = encoder
self.normalize = normalize
self.frontend = frontend
self.specaug = specaug
self.decoder = decoder
self.losses = losses
[docs] def forward(
self,
speech: torch.Tensor,
speech_lengths: torch.Tensor = None,
label: torch.Tensor = None,
**kwargs,
) -> Tuple[torch.Tensor, Dict[str, torch.Tensor], torch.Tensor]:
"""Frontend + Encoder + Decoder + Calc loss
Args:
speech: (Batch, samples)
spk_labels: (Batch, )
kwargs: "utt_id" is among the input.
"""
assert speech.shape[0] == label.shape[0], (speech.shape, label.shape)
batch_size = speech.shape[0]
# 1. Encoder
encoder_out, encoder_out_lens = self.encode(speech, speech_lengths)
# 2. Decoder (baiscally a predction layer after encoder_out)
pred = self.decoder(encoder_out, encoder_out_lens)
if "oc_softmax_loss" in self.losses:
loss = (
self.losses["oc_softmax_loss"](label, encoder_out)
* self.losses["oc_softmax_loss"].weight
)
pred = self.losses["am_softmax_loss"].score(encoder_out)
elif "am_softmax_loss" in self.losses:
loss = (
self.losses["am_softmax_loss"](label, encoder_out)
* self.losses["am_softmax_loss"].weight
)
pred = self.losses["am_softmax_loss"].score(encoder_out)
else:
loss = (
self.losses["binary_loss"](pred, label)
* self.losses["binary_loss"].weight
)
acc = torch.sum(((pred.view(-1) > 0.5) == (label.view(-1) > 0.5))) / batch_size
stats = dict(
loss=loss.detach(),
acc=acc.detach(),
)
loss, stats, weight = force_gatherable((loss, stats, batch_size), loss.device)
return loss, stats, weight
[docs] def collect_feats(
self,
speech: torch.Tensor,
speech_lengths: torch.Tensor,
**kwargs,
) -> Dict[str, torch.Tensor]:
feats, feats_lengths = self._extract_feats(speech, speech_lengths)
return {"feats": feats, "feats_lengths": feats_lengths}
[docs] def encode(
self,
speech: torch.Tensor,
speech_lengths: torch.Tensor,
) -> Tuple[torch.Tensor, torch.Tensor]:
"""Frontend + Encoder
Args:
speech: (Batch, Length, ...)
speech_lengths: (Batch,)
bottleneck_feats: (Batch, Length, ...): used for enh + diar
"""
with autocast(False):
# 1. Extract feats
feats, feats_lengths = self._extract_feats(speech, speech_lengths)
# 2. Data augmentation
if self.specaug is not None and self.training:
feats, feats_lengths = self.specaug(feats, feats_lengths)
# 3. Normalization for feature: e.g. Global-CMVN, Utterance-CMVN
if self.normalize is not None:
feats, feats_lengths = self.normalize(feats, feats_lengths)
# Pre-encoder, e.g. used for raw input data
if self.preencoder is not None:
feats, feats_lengths = self.preencoder(feats, feats_lengths)
# 4. Forward encoder
# feats: (Batch, Length, Dim)
# -> encoder_out: (Batch, Length2, Dim)
encoder_out, encoder_out_lens, _ = self.encoder(feats, feats_lengths)
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(),
)
return encoder_out, encoder_out_lens
def _extract_feats(
self, speech: torch.Tensor, speech_lengths: torch.Tensor
) -> Tuple[torch.Tensor, torch.Tensor]:
batch_size = speech.shape[0]
speech_lengths = (
speech_lengths
if speech_lengths is not None
else torch.ones(batch_size).int() * speech.shape[1]
)
assert speech_lengths.dim() == 1, speech_lengths.shape
# for data-parallel
speech = speech[:, : speech_lengths.max()]
if self.frontend is not None:
# Frontend
# e.g. STFT and Feature extract
# data_loader may send time-domain signal in this case
# speech (Batch, NSamples) -> feats: (Batch, NFrames, Dim)
feats, feats_lengths = self.frontend(speech, speech_lengths)
else:
# No frontend and no feature extract
feats, feats_lengths = speech, speech_lengths
return feats, feats_lengths