def __init__(self, n_loop, n_layer, a_channels, r_channels, s_channels, use_embed_tanh):
super().__init__()
self.save_hyperparameters()
self.encoder = UpsampleNet(
n_loop*n_layer,
r_channels)
self.decoder = WaveNet(
n_loop,
n_layer,
a_channels,
r_channels,
s_channels,
use_embed_tanh)
use_gpu = True
chainer.cuda.get_device_from_id(args.gpu).use()
else:
use_gpu = False
# Preprocess
_, condition, _ = Preprocess(
sr=16000, n_fft=1024, hop_length=256, n_mels=128, top_db=20,
length=None, quantize=args.a_channels)(args.input)
x = numpy.zeros([1, args.a_channels, 1, 1], dtype=condition.dtype)
condition = numpy.expand_dims(condition, axis=0)
# Define networks
encoder = UpsampleNet(args.n_loop * args.n_layer, args.r_channels)
decoder = WaveNet(
args.n_loop, args.n_layer,
args.a_channels, args.r_channels, args.s_channels,
args.use_embed_tanh)
# Load trained parameters
chainer.serializers.load_npz(
args.model, encoder, 'updater/model:main/predictor/encoder/')
chainer.serializers.load_npz(
args.model, decoder, 'updater/model:main/predictor/decoder/')
# Non-autoregressive generate
if use_gpu:
x = chainer.cuda.to_gpu(x, device=args.gpu)
condition = chainer.cuda.to_gpu(condition, device=args.gpu)
encoder.to_gpu(device=args.gpu)
decoder.to_gpu(device=args.gpu)
x = chainer.Variable(x)
# Datasets
if not os.path.isdir(args.dataset):
raise RuntimeError('Dataset directory not found: {}'.format(args.dataset))
paths = sorted([
str(path) for path in pathlib.Path(args.dataset).glob('wav48/*/*.wav')])
preprocess = Preprocess(
sr=16000, n_fft=1024, hop_length=256, n_mels=128, top_db=20,
length=args.length, quantize=args.a_channels)
dataset = chainer.datasets.TransformDataset(paths, preprocess)
train, valid = chainer.datasets.split_dataset_random(
dataset, int(len(dataset) * 0.9), args.seed)
# Networks
encoder = UpsampleNet(args.n_loop * args.n_layer, args.r_channels)
decoder = WaveNet(
args.n_loop, args.n_layer,
args.a_channels, args.r_channels, args.s_channels,
args.use_embed_tanh)
model = chainer.links.Classifier(EncoderDecoderModel(encoder, decoder))
# Optimizer
optimizer = chainer.optimizers.Adam(1e-4)
optimizer.setup(model)
# Iterators
train_iter = chainer.iterators.MultiprocessIterator(
train, args.batchsize,
n_processes=args.process, n_prefetch=args.prefetch)
valid_iter = chainer.iterators.MultiprocessIterator(
valid, args.batchsize, repeat=False, shuffle=False,
n_processes=args.process, n_prefetch=args.prefetch)
chainer.global_config.autotune = True
# Preprocess
_, condition, _ = Preprocess(sr=16000,
n_fft=1024,
hop_length=256,
n_mels=128,
top_db=20,
length=None,
quantize=args.a_channels)(args.input)
x = numpy.zeros([1, args.a_channels, 1, 1], dtype=condition.dtype)
condition = numpy.expand_dims(condition, axis=0)
# Define networks
encoder = UpsampleNet(args.n_loop * args.n_layer, args.r_channels)
decoder = WaveNet(args.n_loop, args.n_layer, args.a_channels, args.r_channels,
args.s_channels, args.use_embed_tanh)
# Load trained parameters
chainer.serializers.load_npz(args.model, encoder,
'updater/model:main/predictor/encoder/')
chainer.serializers.load_npz(args.model, decoder,
'updater/model:main/predictor/decoder/')
# Non-autoregressive generate
x = device.send(x)
condition = device.send(condition)
encoder.to_device(device)
decoder.to_device(device)
x = chainer.Variable(x)
condition = chainer.Variable(condition)
conditions = encoder(condition)
device.use()
if device.xp is chainer.backends.cuda.cupy:
chainer.global_config.autotune = True
# Preprocess
_, condition, _ = Preprocess(
sr=16000, n_fft=1024, hop_length=256, n_mels=128, top_db=20,
length=None, quantize=args.a_channels)(args.input)
x = numpy.zeros([1, args.a_channels, 1, 1], dtype=condition.dtype)
condition = numpy.expand_dims(condition, axis=0)
# Define networks
encoder = UpsampleNet(args.n_loop * args.n_layer, args.r_channels)
decoder = WaveNet(
args.n_loop, args.n_layer,
args.a_channels, args.r_channels, args.s_channels,
args.use_embed_tanh)
# Load trained parameters
chainer.serializers.load_npz(
args.model, encoder, 'updater/model:main/predictor/encoder/')
chainer.serializers.load_npz(
args.model, decoder, 'updater/model:main/predictor/decoder/')
# Non-autoregressive generate
x = device.send(x)
condition = device.send(condition)
encoder.to_device(device)
decoder.to_device(device)
x = chainer.Variable(x)
condition = chainer.Variable(condition)
yield output, target
if __name__ == '__main__':
args = parseInput()
prefix = args.event + args.channel + str(args.freq)
fileCheck(prefix)
tf.compat.v1.logging.set_verbosity(tf.compat.v1.logging.ERROR)
gpus = tf.config.experimental.list_physical_devices('GPU')
print('number of gpus: ', len(gpus))
if args.save_file:
stdoutOrigin = sys.stdout
sys.stdout = open(args.output + '.txt', 'w')
model = WaveNet(args.num_residuals, args.num_filters)
if args.model_path is not None:
model.load_weights(args.model_path)
model.build((None, 8192, 1))
optimizer = tf.keras.optimizers.Adam(learning_rate=args.learning_rate)
criteria = tf.keras.losses.MeanSquaredError()
losses = []
maxSNR = np.linspace(1.75, 1.0, args.epoch)
for epoch in range(args.epoch):
train_dataset = tf.data.Dataset.from_generator(
generator, (tf.float64, tf.float64), ((8192, 1), 8192),
(args.train_file, prefix, args.blank_ratio, (0.5, maxSNR[epoch])))
train_dataset = train_dataset.shuffle(buffer_size=9861).batch(
args.batch_size)
for (batch_n, (input, target)) in enumerate(train_dataset):