def __init__(self,
num_chars,
num_speakers,
r,
attn_win=False,
attn_norm="softmax",
prenet_type="original",
prenet_dropout=True,
forward_attn=False,
trans_agent=False,
forward_attn_mask=False,
location_attn=True,
separate_stopnet=True):
super(Tacotron2, self).__init__()
self.n_mel_channels = 80
self.n_frames_per_step = r
self.embedding = nn.Embedding(num_chars, 512)
std = sqrt(2.0 / (num_chars + 512))
val = sqrt(3.0) * std # uniform bounds for std
self.embedding.weight.data.uniform_(-val, val)
if num_speakers > 1:
self.speaker_embedding = nn.Embedding(num_speakers, 512)
self.speaker_embedding.weight.data.normal_(0, 0.3)
self.encoder = Encoder(512)
self.decoder = Decoder(512, self.n_mel_channels, r, attn_win,
attn_norm, prenet_type, prenet_dropout,
forward_attn, trans_agent, forward_attn_mask,
location_attn, separate_stopnet)
self.postnet = Postnet(self.n_mel_channels)
def __init__(self,
num_chars,
num_speakers,
r,
postnet_output_dim=80,
decoder_output_dim=80,
attn_type='original',
attn_win=False,
attn_norm="softmax",
prenet_type="original",
prenet_dropout=True,
forward_attn=False,
trans_agent=False,
forward_attn_mask=False,
location_attn=True,
attn_K=5,
separate_stopnet=True,
bidirectional_decoder=False,
gst=False):
super(Tacotron2, self).__init__()
self.postnet_output_dim = postnet_output_dim
self.decoder_output_dim = decoder_output_dim
self.gst = gst
self.n_frames_per_step = r
self.bidirectional_decoder = bidirectional_decoder
decoder_dim = 512 if num_speakers > 1 else 512
encoder_dim = 512 if num_speakers > 1 else 512
proj_speaker_dim = 80 if num_speakers > 1 else 0
# embedding layer
self.embedding = nn.Embedding(num_chars, 512, padding_idx=0)
std = sqrt(2.0 / (num_chars + 512))
val = sqrt(3.0) * std # uniform bounds for std
self.embedding.weight.data.uniform_(-val, val)
if num_speakers > 1:
self.speaker_embedding = nn.Embedding(num_speakers, 512)
self.speaker_embedding.weight.data.normal_(0, 0.3)
self.speaker_embeddings = None
self.speaker_embeddings_projected = None
self.encoder = Encoder(encoder_dim)
self.decoder = Decoder(decoder_dim, self.decoder_output_dim, r, attn_type, attn_win,
attn_norm, prenet_type, prenet_dropout,
forward_attn, trans_agent, forward_attn_mask,
location_attn, attn_K, separate_stopnet, proj_speaker_dim)
if self.bidirectional_decoder:
self.decoder_backward = copy.deepcopy(self.decoder)
self.postnet = Postnet(self.postnet_output_dim)
# global style token layers
if self.gst:
print('running with GST')
gst_embedding_dim = encoder_dim
self.gst_layer = GST(num_mel=80,
num_heads=4,
num_style_tokens=10,
embedding_dim=gst_embedding_dim)
def __init__(self,
num_chars,
num_speakers,
r,
postnet_output_dim=80,
decoder_output_dim=80,
attn_type='original',
attn_win=False,
attn_norm="softmax",
prenet_type="original",
prenet_dropout=True,
forward_attn=False,
trans_agent=False,
forward_attn_mask=False,
location_attn=True,
attn_K=5,
separate_stopnet=True,
bidirectional_decoder=False,
stop_token=0.7,
VAE_params=None ,):
super(Tacotron2, self).__init__()
self.VAE_params = VAE_params
self.stop_token=stop_token
self.postnet_output_dim = postnet_output_dim
self.decoder_output_dim = decoder_output_dim
self.n_frames_per_step = r
self.bidirectional_decoder = bidirectional_decoder
decoder_dim = 512 if num_speakers > 1 else 512
encoder_dim = 512 if num_speakers > 1 else 512
proj_speaker_dim = 80 if num_speakers > 1 else 0
# embedding layer
self.embedding = nn.Embedding(num_chars, 512)
std = sqrt(2.0 / (num_chars + 512))
val = sqrt(3.0) * std # uniform bounds for std
self.embedding.weight.data.uniform_(-val, val)
if num_speakers > 1:
self.speaker_embedding = nn.Embedding(num_speakers, 512)
self.speaker_embedding.weight.data.normal_(0, 0.3)
self.speaker_embeddings = None
self.speaker_embeddings_projected = None
self.encoder = Encoder(encoder_dim)
self.decoder = Decoder(decoder_dim, self.decoder_output_dim, r, attn_type, attn_win,
attn_norm, prenet_type, prenet_dropout,
forward_attn, trans_agent, forward_attn_mask,
location_attn, attn_K, separate_stopnet, proj_speaker_dim,self.stop_token)
if self.bidirectional_decoder:
self.decoder_backward = copy.deepcopy(self.decoder)
self.postnet = Postnet(self.postnet_output_dim)
self.vae_gst = VAE_GST(VAE_params)
def __init__(self,
num_chars,
r,
attn_win=False,
attn_norm="softmax",
prenet_type="original",
forward_attn=False,
trans_agent=False):
super(Tacotron2, self).__init__()
self.n_mel_channels = 80
self.n_frames_per_step = r
self.embedding = nn.Embedding(num_chars, 512)
std = sqrt(2.0 / (num_chars + 512))
val = sqrt(3.0) * std # uniform bounds for std
self.embedding.weight.data.uniform_(-val, val)
self.encoder = Encoder(512)
self.decoder = Decoder(512, self.n_mel_channels, r, attn_win,
attn_norm, prenet_type, forward_attn,
trans_agent)
self.postnet = Postnet(self.n_mel_channels)
def __init__(self,
num_chars,
num_speakers,
r,
postnet_output_dim=80,
decoder_output_dim=80,
attn_type='original',
attn_win=False,
attn_norm="softmax",
prenet_type="original",
prenet_dropout=True,
forward_attn=False,
trans_agent=False,
forward_attn_mask=False,
location_attn=True,
attn_K=5,
separate_stopnet=True,
bidirectional_decoder=False,
double_decoder_consistency=False,
ddc_r=None,
gst=False,
gst_embedding_dim=512,
gst_num_heads=4,
gst_style_tokens=10):
super(Tacotron2,
self).__init__(num_chars, num_speakers, r, postnet_output_dim,
decoder_output_dim, attn_type, attn_win,
attn_norm, prenet_type, prenet_dropout,
forward_attn, trans_agent, forward_attn_mask,
location_attn, attn_K, separate_stopnet,
bidirectional_decoder, double_decoder_consistency,
ddc_r, gst)
# init layer dims
speaker_embedding_dim = 512 if num_speakers > 1 else 0
gst_embedding_dim = gst_embedding_dim if self.gst else 0
decoder_in_features = 512 + speaker_embedding_dim + gst_embedding_dim
encoder_in_features = 512 if num_speakers > 1 else 512
proj_speaker_dim = 80 if num_speakers > 1 else 0
# embedding layer
self.embedding = nn.Embedding(num_chars, 512, padding_idx=0)
std = sqrt(2.0 / (num_chars + 512))
val = sqrt(3.0) * std # uniform bounds for std
self.embedding.weight.data.uniform_(-val, val)
# speaker embedding layer
if num_speakers > 1:
self.speaker_embedding = nn.Embedding(num_speakers,
speaker_embedding_dim)
self.speaker_embedding.weight.data.normal_(0, 0.3)
self.encoder = Encoder(encoder_in_features)
self.decoder = Decoder(decoder_in_features, self.decoder_output_dim, r,
attn_type, attn_win, attn_norm, prenet_type,
prenet_dropout, forward_attn, trans_agent,
forward_attn_mask, location_attn, attn_K,
separate_stopnet, proj_speaker_dim)
self.postnet = Postnet(self.postnet_output_dim)
# global style token layers
if self.gst:
self.gst_layer = GST(num_mel=80,
num_heads=gst_num_heads,
num_style_tokens=gst_style_tokens,
embedding_dim=gst_embedding_dim)
# backward pass decoder
if self.bidirectional_decoder:
self._init_backward_decoder()
# setup DDC
if self.double_decoder_consistency:
self.coarse_decoder = Decoder(
decoder_in_features, self.decoder_output_dim, ddc_r, attn_type,
attn_win, attn_norm, prenet_type, prenet_dropout, forward_attn,
trans_agent, forward_attn_mask, location_attn, attn_K,
separate_stopnet, proj_speaker_dim)
class Tacotron2(TacotronAbstract):
def __init__(self,
num_chars,
num_speakers,
r,
postnet_output_dim=80,
decoder_output_dim=80,
attn_type='original',
attn_win=False,
attn_norm="softmax",
prenet_type="original",
prenet_dropout=True,
forward_attn=False,
trans_agent=False,
forward_attn_mask=False,
location_attn=True,
attn_K=5,
separate_stopnet=True,
bidirectional_decoder=False,
double_decoder_consistency=False,
ddc_r=None,
gst=False,
gst_embedding_dim=512,
gst_num_heads=4,
gst_style_tokens=10):
super(Tacotron2,
self).__init__(num_chars, num_speakers, r, postnet_output_dim,
decoder_output_dim, attn_type, attn_win,
attn_norm, prenet_type, prenet_dropout,
forward_attn, trans_agent, forward_attn_mask,
location_attn, attn_K, separate_stopnet,
bidirectional_decoder, double_decoder_consistency,
ddc_r, gst)
# init layer dims
speaker_embedding_dim = 512 if num_speakers > 1 else 0
gst_embedding_dim = gst_embedding_dim if self.gst else 0
decoder_in_features = 512 + speaker_embedding_dim + gst_embedding_dim
encoder_in_features = 512 if num_speakers > 1 else 512
proj_speaker_dim = 80 if num_speakers > 1 else 0
# embedding layer
self.embedding = nn.Embedding(num_chars, 512, padding_idx=0)
std = sqrt(2.0 / (num_chars + 512))
val = sqrt(3.0) * std # uniform bounds for std
self.embedding.weight.data.uniform_(-val, val)
# speaker embedding layer
if num_speakers > 1:
self.speaker_embedding = nn.Embedding(num_speakers,
speaker_embedding_dim)
self.speaker_embedding.weight.data.normal_(0, 0.3)
self.encoder = Encoder(encoder_in_features)
self.decoder = Decoder(decoder_in_features, self.decoder_output_dim, r,
attn_type, attn_win, attn_norm, prenet_type,
prenet_dropout, forward_attn, trans_agent,
forward_attn_mask, location_attn, attn_K,
separate_stopnet, proj_speaker_dim)
self.postnet = Postnet(self.postnet_output_dim)
# global style token layers
if self.gst:
self.gst_layer = GST(num_mel=80,
num_heads=gst_num_heads,
num_style_tokens=gst_style_tokens,
embedding_dim=gst_embedding_dim)
# backward pass decoder
if self.bidirectional_decoder:
self._init_backward_decoder()
# setup DDC
if self.double_decoder_consistency:
self.coarse_decoder = Decoder(
decoder_in_features, self.decoder_output_dim, ddc_r, attn_type,
attn_win, attn_norm, prenet_type, prenet_dropout, forward_attn,
trans_agent, forward_attn_mask, location_attn, attn_K,
separate_stopnet, proj_speaker_dim)
@staticmethod
def shape_outputs(mel_outputs, mel_outputs_postnet, alignments):
mel_outputs = mel_outputs.transpose(1, 2)
mel_outputs_postnet = mel_outputs_postnet.transpose(1, 2)
return mel_outputs, mel_outputs_postnet, alignments
def compute_gst(self, inputs, style_input):
""" Compute global style token """
device = inputs.device
if isinstance(style_input, dict):
query = torch.zeros(1, 1, self.gst_embedding_dim // 2).to(device)
_GST = torch.tanh(self.gst_layer.style_token_layer.style_tokens)
gst_outputs = torch.zeros(1, 1, self.gst_embedding_dim).to(device)
for k_token, v_amplifier in style_input.items():
key = _GST[int(k_token)].unsqueeze(0).expand(1, -1, -1)
gst_outputs_att = self.gst_layer.style_token_layer.attention(
query, key)
gst_outputs = gst_outputs + gst_outputs_att * v_amplifier
elif style_input is None:
gst_outputs = torch.zeros(1, 1, self.gst_embedding_dim).to(device)
else:
gst_outputs = self.gst_layer(style_input) # pylint: disable=not-callable
embedded_gst = gst_outputs.repeat(1, inputs.size(1), 1)
return inputs, embedded_gst
def forward(self,
text,
text_lengths,
mel_specs=None,
mel_lengths=None,
speaker_ids=None):
# compute mask for padding
# B x T_in_max (boolean)
input_mask, output_mask = self.compute_masks(text_lengths, mel_lengths)
# B x D_embed x T_in_max
embedded_inputs = self.embedding(text).transpose(1, 2)
# B x T_in_max x D_en
encoder_outputs = self.encoder(embedded_inputs, text_lengths)
if self.num_speakers > 1:
embedded_speakers = self.speaker_embedding(speaker_ids)[:, None]
embedded_speakers = embedded_speakers.repeat(
1, encoder_outputs.size(1), 1)
if self.gst:
# B x gst_dim
encoder_outputs, embedded_gst = self.compute_gst(
encoder_outputs, mel_specs)
encoder_outputs = torch.cat(
[encoder_outputs, embedded_gst, embedded_speakers], dim=-1)
else:
encoder_outputs = torch.cat(
[encoder_outputs, embedded_speakers], dim=-1)
else:
if self.gst:
# B x gst_dim
encoder_outputs, embedded_gst = self.compute_gst(
encoder_outputs, mel_specs)
encoder_outputs = torch.cat([encoder_outputs, embedded_gst],
dim=-1)
# encoder_outputs = encoder_outputs * input_mask.unsqueeze(2).expand_as(encoder_outputs)
# B x mel_dim x T_out -- B x T_out//r x T_in -- B x T_out//r
decoder_outputs, alignments, stop_tokens = self.decoder(
encoder_outputs, mel_specs, input_mask)
# # sequence masking
# if mel_lengths is not None:
# decoder_outputs = decoder_outputs * output_mask.unsqueeze(1).expand_as(decoder_outputs)
# B x mel_dim x T_out
postnet_outputs = self.postnet(decoder_outputs)
postnet_outputs = decoder_outputs + postnet_outputs
# # sequence masking
# if output_mask is not None:
# postnet_outputs = postnet_outputs * output_mask.unsqueeze(1).expand_as(postnet_outputs)
# B x T_out x mel_dim -- B x T_out x mel_dim -- B x T_out//r x T_in
decoder_outputs, postnet_outputs, alignments = self.shape_outputs(
decoder_outputs, postnet_outputs, alignments)
if self.bidirectional_decoder:
decoder_outputs_backward, alignments_backward = self._backward_pass(
mel_specs, encoder_outputs, input_mask)
return decoder_outputs, postnet_outputs, alignments, stop_tokens, decoder_outputs_backward, alignments_backward
if self.double_decoder_consistency:
decoder_outputs_backward, alignments_backward = self._coarse_decoder_pass(
mel_specs, encoder_outputs, alignments, input_mask)
return decoder_outputs, postnet_outputs, alignments, stop_tokens, decoder_outputs_backward, alignments_backward
return decoder_outputs, postnet_outputs, alignments, stop_tokens
@torch.no_grad()
def inference(self, text, speaker_ids=None, input_style=None):
embedded_inputs = self.embedding(text).transpose(1, 2)
encoder_outputs = self.encoder.inference(embedded_inputs)
if self.num_speakers > 1:
embedded_speakers = self.speaker_embedding(speaker_ids)[:, None]
embedded_speakers = embedded_speakers.repeat(
1, encoder_outputs.size(1), 1)
if self.gst:
# B x gst_dim
encoder_outputs, embedded_gst = self.compute_gst(
encoder_outputs, input_style)
encoder_outputs = torch.cat(
[encoder_outputs, embedded_gst, embedded_speakers], dim=-1)
else:
encoder_outputs = torch.cat(
[encoder_outputs, embedded_speakers], dim=-1)
else:
if self.gst:
# B x gst_dim
encoder_outputs, embedded_gst = self.compute_gst(
encoder_outputs, input_style)
encoder_outputs = torch.cat([encoder_outputs, embedded_gst],
dim=-1)
decoder_outputs, alignments, stop_tokens = self.decoder.inference(
encoder_outputs)
postnet_outputs = self.postnet(decoder_outputs)
postnet_outputs = decoder_outputs + postnet_outputs
decoder_outputs, postnet_outputs, alignments = self.shape_outputs(
decoder_outputs, postnet_outputs, alignments)
return decoder_outputs, postnet_outputs, alignments, stop_tokens
def inference_truncated(self, text, speaker_ids=None, input_style=None):
"""
Preserve model states for continuous inference
"""
embedded_inputs = self.embedding(text).transpose(1, 2)
encoder_outputs = self.encoder.inference_truncated(embedded_inputs)
if self.num_speakers > 1:
embedded_speakers = self.speaker_embedding(speaker_ids)[:, None]
embedded_speakers = embedded_speakers.repeat(
1, encoder_outputs.size(1), 1)
if self.gst:
# B x gst_dim
encoder_outputs, embedded_gst = self.compute_gst(
encoder_outputs, input_style)
encoder_outputs = torch.cat(
[encoder_outputs, embedded_gst, embedded_speakers], dim=-1)
else:
encoder_outputs = torch.cat(
[encoder_outputs, embedded_speakers], dim=-1)
else:
if self.gst:
# B x gst_dim
encoder_outputs, embedded_gst = self.compute_gst(
encoder_outputs, input_style)
encoder_outputs = torch.cat([encoder_outputs, embedded_gst],
dim=-1)
mel_outputs, alignments, stop_tokens = self.decoder.inference_truncated(
encoder_outputs)
mel_outputs_postnet = self.postnet(mel_outputs)
mel_outputs_postnet = mel_outputs + mel_outputs_postnet
mel_outputs, mel_outputs_postnet, alignments = self.shape_outputs(
mel_outputs, mel_outputs_postnet, alignments)
return mel_outputs, mel_outputs_postnet, alignments, stop_tokens
class Tacotron2(nn.Module):
def __init__(self,
num_chars,
r,
attn_win=False,
attn_norm="softmax",
prenet_type="original",
forward_attn=False,
trans_agent=False):
super(Tacotron2, self).__init__()
self.n_mel_channels = 80
self.n_frames_per_step = r
self.embedding = nn.Embedding(num_chars, 512)
std = sqrt(2.0 / (num_chars + 512))
val = sqrt(3.0) * std # uniform bounds for std
self.embedding.weight.data.uniform_(-val, val)
self.encoder = Encoder(512)
self.decoder = Decoder(512, self.n_mel_channels, r, attn_win,
attn_norm, prenet_type, forward_attn,
trans_agent)
self.postnet = Postnet(self.n_mel_channels)
def shape_outputs(self, mel_outputs, mel_outputs_postnet, alignments):
mel_outputs = mel_outputs.transpose(1, 2)
mel_outputs_postnet = mel_outputs_postnet.transpose(1, 2)
return mel_outputs, mel_outputs_postnet, alignments
def forward(self, text, text_lengths, mel_specs=None):
# compute mask for padding
mask = sequence_mask(text_lengths).to(text.device)
embedded_inputs = self.embedding(text).transpose(1, 2)
encoder_outputs = self.encoder(embedded_inputs, text_lengths)
mel_outputs, stop_tokens, alignments = self.decoder(
encoder_outputs, mel_specs, mask)
mel_outputs_postnet = self.postnet(mel_outputs)
mel_outputs_postnet = mel_outputs + mel_outputs_postnet
mel_outputs, mel_outputs_postnet, alignments = self.shape_outputs(
mel_outputs, mel_outputs_postnet, alignments)
return mel_outputs, mel_outputs_postnet, alignments, stop_tokens
def inference(self, text):
embedded_inputs = self.embedding(text).transpose(1, 2)
encoder_outputs = self.encoder.inference(embedded_inputs)
mel_outputs, stop_tokens, alignments = self.decoder.inference(
encoder_outputs)
mel_outputs_postnet = self.postnet(mel_outputs)
mel_outputs_postnet = mel_outputs + mel_outputs_postnet
mel_outputs, mel_outputs_postnet, alignments = self.shape_outputs(
mel_outputs, mel_outputs_postnet, alignments)
return mel_outputs, mel_outputs_postnet, alignments, stop_tokens
def inference_truncated(self, text):
"""
Preserve model states for continuous inference
"""
embedded_inputs = self.embedding(text).transpose(1, 2)
encoder_outputs = self.encoder.inference_truncated(embedded_inputs)
mel_outputs, stop_tokens, alignments = self.decoder.inference_truncated(
encoder_outputs)
mel_outputs_postnet = self.postnet(mel_outputs)
mel_outputs_postnet = mel_outputs + mel_outputs_postnet
mel_outputs, mel_outputs_postnet, alignments = self.shape_outputs(
mel_outputs, mel_outputs_postnet, alignments)
return mel_outputs, mel_outputs_postnet, alignments, stop_tokens
class Tacotron2(nn.Module):
def __init__(self,
num_chars,
num_speakers,
r,
postnet_output_dim=80,
decoder_output_dim=80,
attn_type='original',
attn_win=False,
attn_norm="softmax",
prenet_type="original",
prenet_dropout=True,
forward_attn=False,
trans_agent=False,
forward_attn_mask=False,
location_attn=True,
attn_K=5,
separate_stopnet=True,
bidirectional_decoder=False):
super(Tacotron2, self).__init__()
self.postnet_output_dim = postnet_output_dim
self.decoder_output_dim = decoder_output_dim
self.n_frames_per_step = r
self.bidirectional_decoder = bidirectional_decoder
decoder_dim = 512 if num_speakers > 1 else 512
encoder_dim = 512 if num_speakers > 1 else 512
proj_speaker_dim = 80 if num_speakers > 1 else 0
# embedding layer
self.embedding = nn.Embedding(num_chars, 512)
std = sqrt(2.0 / (num_chars + 512))
val = sqrt(3.0) * std # uniform bounds for std
self.embedding.weight.data.uniform_(-val, val)
if num_speakers > 1:
self.speaker_embedding = nn.Embedding(num_speakers, 512)
self.speaker_embedding.weight.data.normal_(0, 0.3)
self.speaker_embeddings = None
self.speaker_embeddings_projected = None
self.encoder = Encoder(encoder_dim)
self.decoder = Decoder(decoder_dim, self.decoder_output_dim, r,
attn_type, attn_win, attn_norm, prenet_type,
prenet_dropout, forward_attn, trans_agent,
forward_attn_mask, location_attn, attn_K,
separate_stopnet, proj_speaker_dim)
if self.bidirectional_decoder:
self.decoder_backward = copy.deepcopy(self.decoder)
self.postnet = Postnet(self.postnet_output_dim)
def _init_states(self):
self.speaker_embeddings = None
self.speaker_embeddings_projected = None
@staticmethod
def shape_outputs(mel_outputs, mel_outputs_postnet, alignments):
mel_outputs = mel_outputs.transpose(1, 2)
mel_outputs_postnet = mel_outputs_postnet.transpose(1, 2)
return mel_outputs, mel_outputs_postnet, alignments
def forward(self, text, text_lengths, mel_specs=None, speaker_ids=None):
self._init_states()
# compute mask for padding
mask = sequence_mask(text_lengths).to(text.device)
embedded_inputs = self.embedding(text).transpose(1, 2)
encoder_outputs = self.encoder(embedded_inputs, text_lengths)
encoder_outputs = self._add_speaker_embedding(encoder_outputs,
speaker_ids)
decoder_outputs, alignments, stop_tokens = self.decoder(
encoder_outputs, mel_specs, mask)
postnet_outputs = self.postnet(decoder_outputs)
postnet_outputs = decoder_outputs + postnet_outputs
decoder_outputs, postnet_outputs, alignments = self.shape_outputs(
decoder_outputs, postnet_outputs, alignments)
if self.bidirectional_decoder:
decoder_outputs_backward, alignments_backward = self._backward_inference(
mel_specs, encoder_outputs, mask)
return decoder_outputs, postnet_outputs, alignments, stop_tokens, decoder_outputs_backward, alignments_backward
return decoder_outputs, postnet_outputs, alignments, stop_tokens
def inference(self, text, speaker_ids=None):
embedded_inputs = self.embedding(text).transpose(1, 2)
encoder_outputs = self.encoder.inference(embedded_inputs)
encoder_outputs = self._add_speaker_embedding(encoder_outputs,
speaker_ids)
mel_outputs, alignments, stop_tokens = self.decoder.inference(
encoder_outputs)
mel_outputs_postnet = self.postnet(mel_outputs)
mel_outputs_postnet = mel_outputs + mel_outputs_postnet
mel_outputs, mel_outputs_postnet, alignments = self.shape_outputs(
mel_outputs, mel_outputs_postnet, alignments)
return mel_outputs, mel_outputs_postnet, alignments, stop_tokens
def inference_truncated(self, text, speaker_ids=None):
"""
Preserve model states for continuous inference
"""
embedded_inputs = self.embedding(text).transpose(1, 2)
encoder_outputs = self.encoder.inference_truncated(embedded_inputs)
encoder_outputs = self._add_speaker_embedding(encoder_outputs,
speaker_ids)
mel_outputs, alignments, stop_tokens = self.decoder.inference_truncated(
encoder_outputs)
mel_outputs_postnet = self.postnet(mel_outputs)
mel_outputs_postnet = mel_outputs + mel_outputs_postnet
mel_outputs, mel_outputs_postnet, alignments = self.shape_outputs(
mel_outputs, mel_outputs_postnet, alignments)
return mel_outputs, mel_outputs_postnet, alignments, stop_tokens
def _backward_inference(self, mel_specs, encoder_outputs, mask):
decoder_outputs_b, alignments_b, _ = self.decoder_backward(
encoder_outputs, torch.flip(mel_specs, dims=(1, )), mask,
self.speaker_embeddings_projected)
decoder_outputs_b = decoder_outputs_b.transpose(1, 2)
return decoder_outputs_b, alignments_b
def _add_speaker_embedding(self, encoder_outputs, speaker_ids):
if hasattr(self, "speaker_embedding") and speaker_ids is None:
raise RuntimeError(
" [!] Model has speaker embedding layer but speaker_id is not provided"
)
if hasattr(self, "speaker_embedding") and speaker_ids is not None:
speaker_embeddings = self.speaker_embedding(speaker_ids)
speaker_embeddings.unsqueeze_(1)
speaker_embeddings = speaker_embeddings.expand(
encoder_outputs.size(0), encoder_outputs.size(1), -1)
encoder_outputs = encoder_outputs + speaker_embeddings
return encoder_outputs
class Tacotron2(TacotronAbstract):
def __init__(self,
num_chars,
num_speakers,
r,
postnet_output_dim=80,
decoder_output_dim=80,
attn_type='original',
attn_win=False,
attn_norm="softmax",
prenet_type="original",
prenet_dropout=True,
forward_attn=False,
trans_agent=False,
forward_attn_mask=False,
location_attn=True,
attn_K=5,
separate_stopnet=True,
bidirectional_decoder=False,
double_decoder_consistency=False,
ddc_r=None,
gst=False):
super(Tacotron2,
self).__init__(num_chars, num_speakers, r, postnet_output_dim,
decoder_output_dim, attn_type, attn_win,
attn_norm, prenet_type, prenet_dropout,
forward_attn, trans_agent, forward_attn_mask,
location_attn, attn_K, separate_stopnet,
bidirectional_decoder, double_decoder_consistency,
ddc_r, gst)
decoder_in_features = 512 if num_speakers > 1 else 512
encoder_in_features = 512 if num_speakers > 1 else 512
proj_speaker_dim = 80 if num_speakers > 1 else 0
# base layers
self.embedding = nn.Embedding(num_chars, 512, padding_idx=0)
if num_speakers > 1:
self.speaker_embedding = nn.Embedding(num_speakers, 512)
self.speaker_embedding.weight.data.normal_(0, 0.3)
self.encoder = Encoder(encoder_in_features)
self.decoder = Decoder(decoder_in_features, self.decoder_output_dim, r,
attn_type, attn_win, attn_norm, prenet_type,
prenet_dropout, forward_attn, trans_agent,
forward_attn_mask, location_attn, attn_K,
separate_stopnet, proj_speaker_dim)
self.postnet = Postnet(self.postnet_output_dim)
# global style token layers
if self.gst:
gst_embedding_dim = encoder_in_features
self.gst_layer = GST(num_mel=80,
num_heads=4,
num_style_tokens=10,
embedding_dim=gst_embedding_dim)
# backward pass decoder
if self.bidirectional_decoder:
self._init_backward_decoder()
# setup DDC
if self.double_decoder_consistency:
self.coarse_decoder = Decoder(
decoder_in_features, self.decoder_output_dim, ddc_r, attn_type,
attn_win, attn_norm, prenet_type, prenet_dropout, forward_attn,
trans_agent, forward_attn_mask, location_attn, attn_K,
separate_stopnet, proj_speaker_dim)
@staticmethod
def shape_outputs(mel_outputs, mel_outputs_postnet, alignments):
mel_outputs = mel_outputs.transpose(1, 2)
mel_outputs_postnet = mel_outputs_postnet.transpose(1, 2)
return mel_outputs, mel_outputs_postnet, alignments
def forward(self,
text,
text_lengths,
mel_specs=None,
mel_lengths=None,
speaker_ids=None):
self._init_states()
# compute mask for padding
# B x T_in_max (boolean)
input_mask, output_mask = self.compute_masks(text_lengths, mel_lengths)
# B x D_embed x T_in_max
embedded_inputs = self.embedding(text).transpose(1, 2)
# B x T_in_max x D_en
encoder_outputs = self.encoder(embedded_inputs, text_lengths)
# adding speaker embeddding to encoder output
# TODO: multi-speaker
# B x speaker_embed_dim
if speaker_ids is not None:
self.compute_speaker_embedding(speaker_ids)
if self.num_speakers > 1:
# B x T_in x embed_dim + speaker_embed_dim
encoder_outputs = self._add_speaker_embedding(
encoder_outputs, self.speaker_embeddings)
encoder_outputs = encoder_outputs * input_mask.unsqueeze(2).expand_as(
encoder_outputs)
# global style token
if self.gst:
# B x gst_dim
encoder_outputs = self.compute_gst(encoder_outputs, mel_specs)
# B x mel_dim x T_out -- B x T_out//r x T_in -- B x T_out//r
decoder_outputs, alignments, stop_tokens = self.decoder(
encoder_outputs, mel_specs, input_mask)
# sequence masking
if mel_lengths is not None:
decoder_outputs = decoder_outputs * output_mask.unsqueeze(
1).expand_as(decoder_outputs)
# B x mel_dim x T_out
postnet_outputs = self.postnet(decoder_outputs)
postnet_outputs = decoder_outputs + postnet_outputs
# sequence masking
if output_mask is not None:
postnet_outputs = postnet_outputs * output_mask.unsqueeze(
1).expand_as(postnet_outputs)
# B x T_out x mel_dim -- B x T_out x mel_dim -- B x T_out//r x T_in
decoder_outputs, postnet_outputs, alignments = self.shape_outputs(
decoder_outputs, postnet_outputs, alignments)
if self.bidirectional_decoder:
decoder_outputs_backward, alignments_backward = self._backward_pass(
mel_specs, encoder_outputs, input_mask)
return decoder_outputs, postnet_outputs, alignments, stop_tokens, decoder_outputs_backward, alignments_backward
if self.double_decoder_consistency:
decoder_outputs_backward, alignments_backward = self._coarse_decoder_pass(
mel_specs, encoder_outputs, alignments, input_mask)
return decoder_outputs, postnet_outputs, alignments, stop_tokens, decoder_outputs_backward, alignments_backward
return decoder_outputs, postnet_outputs, alignments, stop_tokens
@torch.no_grad()
def inference(self, text, speaker_ids=None):
embedded_inputs = self.embedding(text).transpose(1, 2)
encoder_outputs = self.encoder.inference(embedded_inputs)
if speaker_ids is not None:
self.compute_speaker_embedding(speaker_ids)
if self.num_speakers > 1:
encoder_outputs = self._add_speaker_embedding(
encoder_outputs, self.speaker_embeddings)
decoder_outputs, alignments, stop_tokens = self.decoder.inference(
encoder_outputs)
postnet_outputs = self.postnet(decoder_outputs)
postnet_outputs = decoder_outputs + postnet_outputs
decoder_outputs, postnet_outputs, alignments = self.shape_outputs(
decoder_outputs, postnet_outputs, alignments)
return decoder_outputs, postnet_outputs, alignments, stop_tokens
def inference_truncated(self, text, speaker_ids=None):
"""
Preserve model states for continuous inference
"""
embedded_inputs = self.embedding(text).transpose(1, 2)
encoder_outputs = self.encoder.inference_truncated(embedded_inputs)
encoder_outputs = self._add_speaker_embedding(encoder_outputs,
speaker_ids)
mel_outputs, alignments, stop_tokens = self.decoder.inference_truncated(
encoder_outputs)
mel_outputs_postnet = self.postnet(mel_outputs)
mel_outputs_postnet = mel_outputs + mel_outputs_postnet
mel_outputs, mel_outputs_postnet, alignments = self.shape_outputs(
mel_outputs, mel_outputs_postnet, alignments)
return mel_outputs, mel_outputs_postnet, alignments, stop_tokens
def _speaker_embedding_pass(self, encoder_outputs, speaker_ids):
# TODO: multi-speaker
# if hasattr(self, "speaker_embedding") and speaker_ids is None:
# raise RuntimeError(" [!] Model has speaker embedding layer but speaker_id is not provided")
# if hasattr(self, "speaker_embedding") and speaker_ids is not None:
# speaker_embeddings = speaker_embeddings.expand(encoder_outputs.size(0),
# encoder_outputs.size(1),
# -1)
# encoder_outputs = encoder_outputs + speaker_embeddings
# return encoder_outputs
pass
class Tacotron2(nn.Module):
def __init__(self,
num_chars,
num_speakers,
r,
attn_win=False,
attn_norm="softmax",
prenet_type="original",
prenet_dropout=True,
forward_attn=False,
trans_agent=False,
forward_attn_mask=False,
location_attn=True,
separate_stopnet=True):
super(Tacotron2, self).__init__()
self.n_mel_channels = 80
self.n_frames_per_step = r
self.embedding = nn.Embedding(num_chars, 512)
std = sqrt(2.0 / (num_chars + 512))
val = sqrt(3.0) * std # uniform bounds for std
self.embedding.weight.data.uniform_(-val, val)
if num_speakers > 1:
self.speaker_embedding = nn.Embedding(num_speakers, 512)
self.speaker_embedding.weight.data.normal_(0, 0.3)
self.encoder = Encoder(512)
self.decoder = Decoder(512, self.n_mel_channels, r, attn_win,
attn_norm, prenet_type, prenet_dropout,
forward_attn, trans_agent, forward_attn_mask,
location_attn, separate_stopnet)
self.postnet = Postnet(self.n_mel_channels)
@staticmethod
def shape_outputs(mel_outputs, mel_outputs_postnet, alignments):
mel_outputs = mel_outputs.transpose(1, 2)
mel_outputs_postnet = mel_outputs_postnet.transpose(1, 2)
return mel_outputs, mel_outputs_postnet, alignments
def forward(self, text, text_lengths, mel_specs=None, speaker_ids=None):
# compute mask for padding
mask = sequence_mask(text_lengths).to(text.device)
embedded_inputs = self.embedding(text).transpose(1, 2)
encoder_outputs = self.encoder(embedded_inputs, text_lengths)
encoder_outputs = self._add_speaker_embedding(encoder_outputs,
speaker_ids)
mel_outputs, stop_tokens, alignments = self.decoder(
encoder_outputs, mel_specs, mask)
mel_outputs_postnet = self.postnet(mel_outputs)
mel_outputs_postnet = mel_outputs + mel_outputs_postnet
mel_outputs, mel_outputs_postnet, alignments = self.shape_outputs(
mel_outputs, mel_outputs_postnet, alignments)
return mel_outputs, mel_outputs_postnet, alignments, stop_tokens
def inference(self, text, speaker_ids=None):
embedded_inputs = self.embedding(text).transpose(1, 2)
encoder_outputs = self.encoder.inference(embedded_inputs)
encoder_outputs = self._add_speaker_embedding(encoder_outputs,
speaker_ids)
mel_outputs, stop_tokens, alignments = self.decoder.inference(
encoder_outputs)
mel_outputs_postnet = self.postnet(mel_outputs)
mel_outputs_postnet = mel_outputs + mel_outputs_postnet
mel_outputs, mel_outputs_postnet, alignments = self.shape_outputs(
mel_outputs, mel_outputs_postnet, alignments)
return mel_outputs, mel_outputs_postnet, alignments, stop_tokens
def inference_truncated(self, text, speaker_ids=None):
"""
Preserve model states for continuous inference
"""
embedded_inputs = self.embedding(text).transpose(1, 2)
encoder_outputs = self.encoder.inference_truncated(embedded_inputs)
encoder_outputs = self._add_speaker_embedding(encoder_outputs,
speaker_ids)
mel_outputs, stop_tokens, alignments = self.decoder.inference_truncated(
encoder_outputs)
mel_outputs_postnet = self.postnet(mel_outputs)
mel_outputs_postnet = mel_outputs + mel_outputs_postnet
mel_outputs, mel_outputs_postnet, alignments = self.shape_outputs(
mel_outputs, mel_outputs_postnet, alignments)
return mel_outputs, mel_outputs_postnet, alignments, stop_tokens
def _add_speaker_embedding(self, encoder_outputs, speaker_ids):
if hasattr(self, "speaker_embedding") and speaker_ids is None:
raise RuntimeError(
" [!] Model has speaker embedding layer but speaker_id is not provided"
)
if hasattr(self, "speaker_embedding") and speaker_ids is not None:
speaker_embeddings = self.speaker_embedding(speaker_ids)
speaker_embeddings.unsqueeze_(1)
speaker_embeddings = speaker_embeddings.expand(
encoder_outputs.size(0), encoder_outputs.size(1), -1)
encoder_outputs = encoder_outputs + speaker_embeddings
return encoder_outputs
