coqui-ai
/
TTS
mirror of https://github.com/coqui-ai/TTS.git
1
0
Fork 0
Code Issues Packages Projects Releases Wiki Activity

Make style

pull/2700/head
Eren G??lge 2023-06-26 17:16:26 +02:00
parent a1c431e6a9
commit c03768bb53
5 changed files with 33 additions and 27 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

2
TTS/tts/configs/fast_speech_config.py
View File

@ -107,7 +107,7 @@ class FastSpeechConfig(BaseTTSConfig):
base_model: str = "forward_tts"
# model specific params
model_args: ForwardTTSArgs = ForwardTTSArgs(use_pitch=False)
model_args: ForwardTTSArgs = field(default_factory=lambda: ForwardTTSArgs(use_pitch=False))
# multi-speaker settings
num_speakers: int = 0

2
TTS/tts/configs/fastspeech2_config.py
View File

@ -123,7 +123,7 @@ class Fastspeech2Config(BaseTTSConfig):
base_model: str = "forward_tts"
# model specific params
model_args: ForwardTTSArgs = ForwardTTSArgs(use_pitch=True, use_energy=True)
model_args: ForwardTTSArgs = field(default_factory=lambda: ForwardTTSArgs(use_pitch=True, use_energy=True))
# multi-speaker settings
num_speakers: int = 0

42
TTS/tts/configs/speedy_speech_config.py
View File

@ -103,26 +103,28 @@ class SpeedySpeechConfig(BaseTTSConfig):
base_model: str = "forward_tts"
# set model args as SpeedySpeech
model_args: ForwardTTSArgs = ForwardTTSArgs(
use_pitch=False,
encoder_type="residual_conv_bn",
encoder_params={
"kernel_size": 4,
"dilations": 4 * [1, 2, 4] + [1],
"num_conv_blocks": 2,
"num_res_blocks": 13,
},
decoder_type="residual_conv_bn",
decoder_params={
"kernel_size": 4,
"dilations": 4 * [1, 2, 4, 8] + [1],
"num_conv_blocks": 2,
"num_res_blocks": 17,
},
out_channels=80,
hidden_channels=128,
positional_encoding=True,
detach_duration_predictor=True,
model_args: ForwardTTSArgs = field(
default_factory=lambda: ForwardTTSArgs(
use_pitch=False,
encoder_type="residual_conv_bn",
encoder_params={
"kernel_size": 4,
"dilations": 4 * [1, 2, 4] + [1],
"num_conv_blocks": 2,
"num_res_blocks": 13,
},
decoder_type="residual_conv_bn",
decoder_params={
"kernel_size": 4,
"dilations": 4 * [1, 2, 4, 8] + [1],
"num_conv_blocks": 2,
"num_res_blocks": 17,
},
out_channels=80,
hidden_channels=128,
positional_encoding=True,
detach_duration_predictor=True,
)
)
# multi-speaker settings

11
TTS/tts/layers/losses.py
View File

@ -165,7 +165,7 @@ class BCELossMasked(nn.Module):
def __init__(self, pos_weight: float = None):
super().__init__()
self.pos_weight = nn.Parameter(torch.tensor([pos_weight]), requires_grad=False)
self.register_buffer("pos_weight", torch.tensor([pos_weight]))
def forward(self, x, target, length):
"""
@ -191,10 +191,15 @@ class BCELossMasked(nn.Module):
mask = sequence_mask(sequence_length=length, max_len=target.size(1))
num_items = mask.sum()
loss = functional.binary_cross_entropy_with_logits(
x.masked_select(mask), target.masked_select(mask), pos_weight=self.pos_weight, reduction="sum"
x.masked_select(mask),
target.masked_select(mask),
pos_weight=self.pos_weight.to(x.device),
reduction="sum",
)
else:
loss = functional.binary_cross_entropy_with_logits(x, target, pos_weight=self.pos_weight, reduction="sum")
loss = functional.binary_cross_entropy_with_logits(
x, target, pos_weight=self.pos_weight.to(x.device), reduction="sum"
)
num_items = torch.numel(x)
loss = loss / num_items
return loss

3
tests/tts_tests/test_tacotron_model.py
View File

@ -16,7 +16,7 @@ from TTS.utils.audio import AudioProcessor
torch.manual_seed(1)
use_cuda = torch.cuda.is_available()
device = torch.device("cuda:0" if torch.cuda.is_available() else "cpu")
device = torch.device("cuda" if use_cuda else "cpu")
config_global = TacotronConfig(num_chars=32, num_speakers=5, out_channels=513, decoder_output_dim=80)
@ -288,7 +288,6 @@ class TacotronCapacitronTrainTest(unittest.TestCase):
batch["text_input"].shape[0], batch["stop_targets"].size(1) // config.r, -1
)
batch["stop_targets"] = (batch["stop_targets"].sum(2) > 0.0).unsqueeze(2).float().squeeze()
model = Tacotron(config).to(device)
criterion = model.get_criterion()
optimizer = model.get_optimizer()