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

add stop token to tacotron testing

This commit is contained in:
Eren Golge 2018-04-30 06:01:02 -07:00
parent 02d50089c4
commit b7e4d214f8
2 changed files with 17 additions and 9 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

12
config.json
View File

@ -12,7 +12,7 @@
"text_cleaner": "english_cleaners", "text_cleaner": "english_cleaners",
"epochs": 50, "epochs": 50,
"lr": 0.004, "lr": 0.002,
"warmup_steps": 4000, "warmup_steps": 4000,
"batch_size": 32, "batch_size": 32,
"eval_batch_size":32, "eval_batch_size":32,
@ -23,14 +23,14 @@
"griffin_lim_iters": 60, "griffin_lim_iters": 60,
"power": 1.2, "power": 1.2,
"dataset": "TWEB", "dataset": "LJSpeech",
"meta_file_train": "transcript_train.txt", "meta_file_train": "metadata_train.csv",
"meta_file_val": "transcript_val.txt", "meta_file_val": "metadata_val.csv",
"data_path": "/data/shared/BibleSpeech/", "data_path": "/data/shared/KeithIto/LJSpeech-1.0/",
"min_seq_len": 0, "min_seq_len": 0,
"num_loader_workers": 8, "num_loader_workers": 8,
"checkpoint": true, "checkpoint": true,
"save_step": 908, "save_step": 600,
"output_path": "/data/shared/erogol_models/" "output_path": "/data/shared/erogol_models/"
} }

14
tests/tacotron_tests.py
View File

@ -5,6 +5,7 @@ import unittest
import numpy as np import numpy as np
from torch import optim from torch import optim
from torch import nn
from TTS.utils.generic_utils import load_config from TTS.utils.generic_utils import load_config
from TTS.layers.losses import L1LossMasked from TTS.layers.losses import L1LossMasked
from TTS.models.tacotron import Tacotron from TTS.models.tacotron import Tacotron
@ -24,7 +25,11 @@ class TacotronTrainTest(unittest.TestCase):
mel_spec = torch.rand(8, 30, c.num_mels).to(device) mel_spec = torch.rand(8, 30, c.num_mels).to(device)
linear_spec = torch.rand(8, 30, c.num_freq).to(device) linear_spec = torch.rand(8, 30, c.num_freq).to(device)
mel_lengths = torch.randint(20, 30, (8,)).long().to(device) mel_lengths = torch.randint(20, 30, (8,)).long().to(device)
stop_targets = torch.zeros(8, 30, 1).float().to(device)
for idx in mel_lengths:
stop_targets[:, int(idx.item()):, 0] = 1.0
criterion = L1LossMasked().to(device) criterion = L1LossMasked().to(device)
criterion_st = nn.BCELoss().to(device)
model = Tacotron(c.embedding_size, model = Tacotron(c.embedding_size,
c.num_freq, c.num_freq,
c.num_mels, c.num_mels,
@ -37,17 +42,20 @@ class TacotronTrainTest(unittest.TestCase):
count += 1 count += 1
optimizer = optim.Adam(model.parameters(), lr=c.lr) optimizer = optim.Adam(model.parameters(), lr=c.lr)
for i in range(5): for i in range(5):
mel_out, linear_out, align = model.forward(input, mel_spec) mel_out, linear_out, align, stop_tokens = model.forward(input, mel_spec)
assert stop_tokens.data.max() <= 1.0
assert stop_tokens.data.min() >= 0.0
optimizer.zero_grad() optimizer.zero_grad()
loss = criterion(mel_out, mel_spec, mel_lengths) loss = criterion(mel_out, mel_spec, mel_lengths)
loss = 0.5 * loss + 0.5 * criterion(linear_out, linear_spec, mel_lengths) stop_loss = criterion_st(stop_tokens, stop_targets)
loss = loss + criterion(linear_out, linear_spec, mel_lengths) + stop_loss
loss.backward() loss.backward()
optimizer.step() optimizer.step()
# check parameter changes # check parameter changes
count = 0 count = 0
for param, param_ref in zip(model.parameters(), model_ref.parameters()): for param, param_ref in zip(model.parameters(), model_ref.parameters()):
# ignore pre-higway layer since it works conditional # ignore pre-higway layer since it works conditional
if count not in [139, 59]: if count not in [141, 59]:
assert (param != param_ref).any(), "param {} with shape {} not updated!! \n{}\n{}".format(count, param.shape, param, param_ref) assert (param != param_ref).any(), "param {} with shape {} not updated!! \n{}\n{}".format(count, param.shape, param, param_ref)
count += 1 count += 1