dhgrs/chainer-Fast-WaveNet

A Chainer implementation of Fast WaveNet(mel-spectrogram vocoder).

chainer-Fast-WaveNet

A Chainer implementation of WaveNet.

Requirements

I trained and generated with

• python(3.5.2)
• chainer(4.0.0b3)
• librosa(0.5.1)

Usage

download dataset

You can download VCTK-Corpus(en) from here. And you can download CMU-ARCTIC(en)/voice-statistics-corpus(ja) very easily via my repository.

set parameters

parameters of training

• batchsize
  • Batch size.
• lr
  • Learning rate.
• ema_mu
  • Rate of exponential moving average. If this is greater than 1 doesn't apply.
• trigger
  • How many times you update the model. You can set this parameter like as (<int>, 'iteration') or (<int>, 'epoch')
• evaluate_interval
  • The interval that you evaluate validation dataset. You can set this parameter like as trigger.
• snapshot_interval
  • The interval that you save snapshot. You can set this parameter like as trigger.
• report_interval
  • The interval that you write log of loss. You can set this parameter like as trigger.

parameters of dataset

• root
  • The root directory of training dataset.
• dataset_type
  • The architecture of the directory of training dataset. Now this parameter supports VCTK, ARCTIC and vs.

parameters of preprocessing

• sr
  • Sampling rate. If it's different from input file, be resampled by librosa.
• n_fft
  • The window size of FFT.
• hop_length
  • The hop length of FFT.
• n_mels
  • The number of mel frequencies.
• top_db
  • The threshold db for triming silence.
• input_dim
  • Input dimension of audio waves. It should be 1 or same as quantize.
• quantize
  • If use_logistic is True it should be 2 ** 16. If False it should be 256.
• length
  • How many samples used for training.
• use_logistic
  • If True use mixture of logistics.

parameters of Encoder(Deconvolution network)

• channels
  • Channels of deconvolution in encoder. The number of elements must be same as upsample_factors.
• upsample_factors
  • The factor of upsampling by deconvolution. The number of elements must be same as channels and the product of elements must be same as hop_length.

parameters of Decoder(WaveNet)

• n_loop
  • If you want to make network like dilations [1, 2, 4, 1, 2, 4] set n_loop as 2.
• n_layer
  • If you want to make network like dilations [1, 2, 4, 1, 2, 4] set n_layer as 3.
• filter_size
  • The filter size of each dilated convolution.
• residual_channels
  • The number of input/output channels of residual blocks.
• dilated_channels
  • The number of output channels of causal dilated convolution layers. This is splited into tanh and sigmoid so the number of hidden units is half of this number.
• skip_channels
  • The number of channels of skip connections and last projection layer.
• n_mixture
  • The number of logistic distribution. It is used only use_logistic is True.
• log_scale_min
  • The number for stability. It is used only use_logistic is True.
• condition_dim
  • The dimension of condition. It must be same as the last element of channels.
• dropout_zero_rate
  • The rate of 0 in dropout. If 0 doesn't apply dropout.

parameters of generating

• use_ema
  • If True use the value of exponential moving average.
• apply_dropout
  • If True apply dropout.

training

(without GPU)
python train.py

(with GPU #n)
python train.py -g n

If you want to use multi GPUs, you can add IDs like below.

python train.py -g 0 1 2

You can resume snapshot and restart training like below.

python train.py -r snapshot_iter_100000

Other arguments -f and -p are parameters for multiprocess in preprocessing. -f means the number of prefetch and -p means the number of processes.

generating

python generate.py -i <input file> -o <output file> -m <trained model>