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

Constant queue size for autoregression window

This commit is contained in:
Eren Golge 2019-02-16 03:18:49 +01:00
parent 11b6080cfd
commit 6ea31e47df
3 changed files with 4 additions and 3 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

2
layers/tacotron.py
View File

@ -309,7 +309,7 @@ class Decoder(nn.Module):
self.memory_size = memory_size if memory_size > 0 else r self.memory_size = memory_size if memory_size > 0 else r
self.memory_dim = memory_dim self.memory_dim = memory_dim
# memory -> |Prenet| -> processed_memory # memory -> |Prenet| -> processed_memory
self.prenet = Prenet(memory_dim * memory_dim * self.memory_size, out_features=[256, 128]) self.prenet = Prenet(memory_dim * self.memory_size, out_features=[256, 128])
# processed_inputs, processed_memory -> |Attention| -> Attention, attention, RNN_State # processed_inputs, processed_memory -> |Attention| -> Attention, attention, RNN_State
self.attention_rnn = AttentionRNNCell( self.attention_rnn = AttentionRNNCell(
out_dim=128, out_dim=128,

3
models/tacotron.py
View File

@ -13,6 +13,7 @@ class Tacotron(nn.Module):
mel_dim=80, mel_dim=80,
r=5, r=5,
padding_idx=None, padding_idx=None,
memory_size=5,
attn_windowing=False): attn_windowing=False):
super(Tacotron, self).__init__() super(Tacotron, self).__init__()
self.r = r self.r = r
@ -23,7 +24,7 @@ class Tacotron(nn.Module):
print(" | > Number of characters : {}".format(num_chars)) print(" | > Number of characters : {}".format(num_chars))
self.embedding.weight.data.normal_(0, 0.3) self.embedding.weight.data.normal_(0, 0.3)
self.encoder = Encoder(embedding_dim) self.encoder = Encoder(embedding_dim)
self.decoder = Decoder(256, mel_dim, r, attn_windowing) self.decoder = Decoder(256, mel_dim, r, memory_size, attn_windowing)
self.postnet = PostCBHG(mel_dim) self.postnet = PostCBHG(mel_dim)
self.last_linear = nn.Sequential( self.last_linear = nn.Sequential(
nn.Linear(self.postnet.cbhg.gru_features * 2, linear_dim), nn.Linear(self.postnet.cbhg.gru_features * 2, linear_dim),

2
train.py
View File

@ -357,7 +357,7 @@ def evaluate(model, criterion, criterion_st, ap, current_step):
def main(args): def main(args):
num_chars = len(phonemes) if c.use_phonemes else len(symbols) num_chars = len(phonemes) if c.use_phonemes else len(symbols)
model = Tacotron(num_chars, c.embedding_size, ap.num_freq, ap.num_mels, c.r) model = Tacotron(num_chars, c.embedding_size, ap.num_freq, ap.num_mels, c.r, c.memory_size)
print(" | > Num output units : {}".format(ap.num_freq), flush=True) print(" | > Num output units : {}".format(ap.num_freq), flush=True)
optimizer = optim.Adam(model.parameters(), lr=c.lr, weight_decay=0) optimizer = optim.Adam(model.parameters(), lr=c.lr, weight_decay=0)