def test_wavernn():
config = WavernnConfig()
config.model_args = WavernnArgs(
rnn_dims=512,
fc_dims=512,
mode="mold",
mulaw=False,
pad=2,
use_aux_net=True,
use_upsample_net=True,
upsample_factors=[4, 8, 8],
feat_dims=80,
compute_dims=128,
res_out_dims=128,
num_res_blocks=10,
)
config.audio.hop_length = 256
config.audio.sample_rate = 2048
dummy_x = torch.rand((2, 1280))
dummy_m = torch.rand((2, 80, 9))
y_size = random.randrange(20, 60)
dummy_y = torch.rand((80, y_size))
# mode: mold
model = Wavernn(config)
output = model(dummy_x, dummy_m)
assert np.all(output.shape == (2, 1280, 30)), output.shape
# mode: gauss
config.model_args.mode = "gauss"
model = Wavernn(config)
output = model(dummy_x, dummy_m)
assert np.all(output.shape == (2, 1280, 2)), output.shape
# mode: quantized
config.model_args.mode = 4
model = Wavernn(config)
output = model(dummy_x, dummy_m)
assert np.all(output.shape == (2, 1280, 2**4)), output.shape
output = model.inference(dummy_y, True, 5500, 550)
assert np.all(output.shape == (256 * (y_size - 1),))
from tests import get_device_id, get_tests_output_path, run_cli
from TTS.vocoder.configs import WavernnConfig
from TTS.vocoder.models.wavernn import WavernnArgs
config_path = os.path.join(get_tests_output_path(), "test_vocoder_config.json")
output_path = os.path.join(get_tests_output_path(), "train_outputs")
config = WavernnConfig(
model_args=WavernnArgs(),
batch_size=8,
eval_batch_size=8,
num_loader_workers=0,
num_eval_loader_workers=0,
run_eval=True,
test_delay_epochs=-1,
epochs=1,
seq_len=256, # for shorter test time
eval_split_size=1,
print_step=1,
print_eval=True,
data_path="tests/data/ljspeech",
output_path=output_path,
)
config.audio.do_trim_silence = True
config.audio.trim_db = 60
config.save_json(config_path)
# train the model for one epoch
command_train = f"CUDA_VISIBLE_DEVICES='{get_device_id()}' python TTS/bin/train_vocoder.py --config_path {config_path} "
run_cli(command_train)
from TTS.utils.audio import AudioProcessor
from TTS.vocoder.configs import WavernnConfig
from TTS.vocoder.datasets.preprocess import load_wav_data
from TTS.vocoder.models.wavernn import Wavernn
output_path = os.path.dirname(os.path.abspath(__file__))
config = WavernnConfig(
batch_size=64,
eval_batch_size=16,
num_loader_workers=4,
num_eval_loader_workers=4,
run_eval=True,
test_delay_epochs=-1,
epochs=10000,
seq_len=1280,
pad_short=2000,
use_noise_augment=False,
eval_split_size=10,
print_step=25,
print_eval=True,
mixed_precision=False,
lr=1e-4,
grad_clip=4,
data_path=os.path.join(output_path, "../LJSpeech-1.1/wavs/"),
output_path=output_path,
)
# init audio processor
ap = AudioProcessor(**config.audio.to_dict())
# load training samples
eval_samples, train_samples = load_wav_data(config.data_path,
import shutil
import numpy as np
from torch.utils.data import DataLoader
from tests import get_tests_output_path, get_tests_path
from TTS.utils.audio import AudioProcessor
from TTS.vocoder.configs import WavernnConfig
from TTS.vocoder.datasets.preprocess import load_wav_feat_data, preprocess_wav_files
from TTS.vocoder.datasets.wavernn_dataset import WaveRNNDataset
file_path = os.path.dirname(os.path.realpath(__file__))
OUTPATH = os.path.join(get_tests_output_path(), "loader_tests/")
os.makedirs(OUTPATH, exist_ok=True)
C = WavernnConfig()
test_data_path = os.path.join(get_tests_path(), "data/ljspeech/")
test_mel_feat_path = os.path.join(test_data_path, "mel")
test_quant_feat_path = os.path.join(test_data_path, "quant")
ok_ljspeech = os.path.exists(test_data_path)
def wavernn_dataset_case(batch_size, seq_len, hop_len, pad, mode, mulaw,
num_workers):
"""run dataloader with given parameters and check conditions"""
ap = AudioProcessor(**C.audio)
C.batch_size = batch_size
C.mode = mode
C.seq_len = seq_len