def main():
"""Run 🐸TTS trainer from terminal. This is also necessary to run DDP training by ```distribute.py```"""
args, config, output_path, _, c_logger, dashboard_logger = init_training(
sys.argv)
trainer = Trainer(args,
config,
output_path,
c_logger,
dashboard_logger,
cudnn_benchmark=False)
trainer.fit()
def main():
try:
args, config, output_path, _, c_logger, dashboard_logger = init_training(
sys.argv)
trainer = Trainer(args, config, output_path, c_logger,
dashboard_logger)
trainer.fit()
except KeyboardInterrupt:
remove_experiment_folder(output_path)
try:
sys.exit(0)
except SystemExit:
os._exit(0) # pylint: disable=protected-access
except Exception: # pylint: disable=broad-except
remove_experiment_folder(output_path)
traceback.print_exc()
sys.exit(1)
import os
from TTS.trainer import Trainer, TrainingArgs, init_training
from TTS.tts.configs import AlignTTSConfig, BaseDatasetConfig
output_path = os.path.dirname(os.path.abspath(__file__))
dataset_config = BaseDatasetConfig(
name="ljspeech", meta_file_train="metadata.csv", path=os.path.join(output_path, "../LJSpeech-1.1/")
)
config = AlignTTSConfig(
batch_size=32,
eval_batch_size=16,
num_loader_workers=4,
num_eval_loader_workers=4,
run_eval=True,
test_delay_epochs=-1,
epochs=1000,
text_cleaner="english_cleaners",
use_phonemes=False,
phoneme_language="en-us",
phoneme_cache_path=os.path.join(output_path, "phoneme_cache"),
print_step=25,
print_eval=True,
mixed_precision=False,
output_path=output_path,
datasets=[dataset_config],
)
args, config, output_path, _, c_logger, dashboard_logger = init_training(TrainingArgs(), config)
trainer = Trainer(args, config, output_path, c_logger, dashboard_logger)
trainer.fit()
spec_gain=1.0,
signal_norm=False,
do_amp_to_db_linear=False,
)
config = VitsConfig(
audio=audio_config,
run_name="vits_ljspeech",
batch_size=48,
eval_batch_size=16,
batch_group_size=5,
num_loader_workers=4,
num_eval_loader_workers=4,
run_eval=True,
test_delay_epochs=-1,
epochs=1000,
text_cleaner="english_cleaners",
use_phonemes=True,
phoneme_language="en-us",
phoneme_cache_path=os.path.join(output_path, "phoneme_cache"),
compute_input_seq_cache=True,
print_step=25,
print_eval=True,
mixed_precision=True,
max_seq_len=500000,
output_path=output_path,
datasets=[dataset_config],
)
args, config, output_path, _, c_logger, tb_logger = init_training(TrainingArgs(), config)
trainer = Trainer(args, config, output_path, c_logger, tb_logger, cudnn_benchmark=True)
trainer.fit()
