wavenet = WaveNet(params)
else:
params = Params()
params.audio_channels = 256
params.causal_conv_no_bias = True
params.causal_conv_kernel_width = 2
params.causal_conv_channels = [128]
params.residual_conv_dilation_no_bias = True
params.residual_conv_projection_no_bias = True
params.residual_conv_kernel_width = 2
params.residual_conv_channels = [32, 32, 32, 32, 32, 32, 32, 32, 32]
params.residual_num_blocks = 5
params.softmax_conv_no_bias = True
params.softmax_conv_kernel_width = 2
params.softmax_conv_channels = [128, 256]
params.learning_rate = 0.1
params.gradient_momentum = 0.9
params.weight_decay = 0.000001
params.gradient_clipping = 10.0
params.gpu_enabled = True if args.gpu_enabled == 1 else False
if args.use_faster_wavenet:
wavenet = FasterWaveNet(params)
else:
wavenet = WaveNet(params)
wavenet = FasterWaveNet(params) else: params = Params() params.gpu_enabled = True if args.gpu_enabled == 1 else False params.quantization_steps = 256 params.sampling_rate = 8000 params.causal_conv_no_bias = False params.causal_conv_filter_width = 2 params.causal_conv_channels = [128] params.residual_conv_dilation_no_bias = True params.residual_conv_projection_no_bias = True params.residual_conv_filter_width = 3 params.residual_conv_channels = [32, 32, 32, 32, 32, 32] params.residual_num_blocks = 2 params.softmax_conv_no_bias = False params.softmax_conv_channels = [128, 256] params.learning_rate = 0.001 params.gradient_momentum = 0.9 params.weight_decay = 0.000001 params.gradient_clipping = 10.0 wavenet = FasterWaveNet(params) f = open(filename, "w") json.dump(params.to_dict(), f, indent=4) params.dump() wavenet.load(args.model_dir)