import json, os
from args import args
from wavenet import WaveNet, Params
from faster_wavenet import FasterWaveNet
# load params.json
try:
os.mkdir(args.params_dir)
except:
pass
filename = args.params_dir + "/{}".format(args.params_filename)
if os.path.isfile(filename):
f = open(filename)
try:
dict = json.load(f)
params = Params(dict)
except:
raise Exception("could not load {}".format(filename))
params.gpu_enabled = True if args.gpu_enabled == 1 else False
if args.use_faster_wavenet:
wavenet = FasterWaveNet(params)
else:
wavenet = WaveNet(params)
else:
params = Params()
params.audio_channels = 256
params.causal_conv_no_bias = True
params.causal_conv_kernel_width = 2
# -*- coding: utf-8 -*-
import json, os
from args import args
from wavenet import WaveNet, Params
# load params.json
try:
os.mkdir(args.params_dir)
except:
pass
filename = args.params_dir + "/{}".format(args.params_filename)
if os.path.isfile(filename):
f = open(filename)
dict = json.load(f)
params = Params(dict)
wavenet = WaveNet(params)
else:
params = Params()
params.quantization_steps = 6
params.causal_conv_no_bias = True
params.causal_conv_filter_width = 2
params.causal_conv_channels = [4]
params.residual_conv_dilation_no_bias = True
params.residual_conv_projection_no_bias = True
params.residual_conv_filter_width = 2
params.residual_conv_channels = [3, 3]
params.residual_num_blocks = 1
params.softmax_conv_no_bias = False
# -*- coding: utf-8 -*-
import json, os
from args import args
from chainer import cuda
from wavenet import WaveNet, Params
# load params.json
try:
os.mkdir(args.model_dir)
except:
pass
filename = args.model_dir + "/wavenet.json"
if os.path.isfile(filename):
f = open(filename)
dict = json.load(f)
params = Params(dict)
wavenet = WaveNet(params)
else:
params = Params()
params.quantization_steps = 10
params.causal_conv_no_bias = True
params.causal_conv_filter_width = 3
params.causal_conv_channels = [32]
params.residual_conv_dilation_no_bias = True
params.residual_conv_projection_no_bias = True
params.residual_conv_filter_width = 3
params.residual_conv_channels = [16, 16]
params.residual_num_blocks = 1
# -*- coding: utf-8 -*-
import json, os
from args import args
from wavenet import WaveNet, Params
# load params.json
try:
os.mkdir(args.params_dir)
except:
pass
filename = args.params_dir + "/{}".format(args.params_filename)
if os.path.isfile(filename):
f = open(filename)
dict = json.load(f)
params = Params(dict)
wavenet = WaveNet(params)
else:
params = Params()
params.gpu_enabled = True if args.gpu_enabled == 1 else False
params.audio_channels = 3
params.residual_conv_kernel_width = 2
params.residual_conv_channels = [3]
params.softmax_conv_channels = [3]
params.causal_conv_channels = [3, 3, 3]
params.residual_conv_dilations = [2]
params.causal_conv_apply_batchnorm = False
params.residual_conv_apply_batchnorm = False
params.softmax_conv_apply_batchnorm = False
wavenet = WaveNet(params)
f = open(filename, "w")
from chainer import cuda
from wavenet import WaveNet, Params
from faster_wavenet import FasterWaveNet
# load params.json
try:
os.mkdir(args.model_dir)
except:
pass
filename = args.model_dir + "/wavenet.json"
if os.path.isfile(filename):
print "loading", filename
f = open(filename)
try:
dict = json.load(f)
params = Params(dict)
except:
raise Exception("could not load {}".format(filename))
else:
params = Params()
params.quantization_steps = 256
params.sampling_rate = 8000
params.causal_conv_no_bias = True
params.causal_conv_filter_width = 2
params.causal_conv_channels = [256]
params.residual_conv_dilation_no_bias = True
params.residual_conv_projection_no_bias = True
params.residual_conv_filter_width = 2
params.residual_conv_channels = [